CN117055601B - Unmanned aerial vehicle meal delivery path planning method, unmanned aerial vehicle meal delivery path planning device, unmanned aerial vehicle meal delivery path planning equipment and storage medium - Google Patents
Unmanned aerial vehicle meal delivery path planning method, unmanned aerial vehicle meal delivery path planning device, unmanned aerial vehicle meal delivery path planning equipment and storage medium Download PDFInfo
- Publication number
- CN117055601B CN117055601B CN202311136458.5A CN202311136458A CN117055601B CN 117055601 B CN117055601 B CN 117055601B CN 202311136458 A CN202311136458 A CN 202311136458A CN 117055601 B CN117055601 B CN 117055601B
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- unmanned aerial
- aerial vehicle
- module
- path
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 235000012054 meals Nutrition 0.000 title claims abstract description 49
- 238000003860 storage Methods 0.000 title claims abstract description 12
- 230000008859 change Effects 0.000 claims description 44
- 230000001133 acceleration Effects 0.000 claims description 36
- 238000004364 calculation method Methods 0.000 claims description 31
- 238000006073 displacement reaction Methods 0.000 claims description 17
- 238000001514 detection method Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 9
- 238000013459 approach Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 7
- 238000005457 optimization Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- 230000002787 reinforcement Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 230000008447 perception Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000024703 flight behavior Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method, a device, equipment and a storage medium for planning a meal delivery path of an unmanned aerial vehicle, which are used for solving the technical problems that the conventional unmanned aerial vehicle is poor in meal delivery path planning and easy to collide. The invention comprises the following steps: acquiring scene information of an unmanned aerial vehicle flight environment, and acquiring pedestrian position information from the scene information; generating a pedestrian prediction track according to the pedestrian position information; generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track; optimizing the initial flight path to obtain a target flight path; and planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters.
Description
Technical Field
The present invention relates to the field of path planning technologies, and in particular, to a method, an apparatus, a device, and a storage medium for planning a meal delivery path of an unmanned aerial vehicle.
Background
In an autonomous meal delivery task, path planning in a dynamic pedestrian environment is a challenging problem, and not only the obstacle avoidance problem of a static obstacle, but also the factors such as the actions of pedestrians, obstacle avoidance, path safety and the like are required to be considered. The path planning method based on the conventional algorithm is one of the most basic methods. Among them, dijkstra's algorithm is the most common path planning algorithm for static environments.
However, in a dynamic pedestrian environment, there are some limitations to the application of these conventional methods because they cannot accommodate the movement and obstacle avoidance needs of pedestrians. Therefore, a path planning method based on machine learning is attracting attention gradually for a dynamic pedestrian environment. Reinforcement learning is one of the widely used methods. Through regarding unmanned aerial vehicle as the agent, let it carry out interactive study with the environment, can make it optimize route planning strategy step by step. Reinforcement Learning methods such as Q-Learning and deep reinforcement Learning have been used for path planning problems with some benefit. In addition, the hybrid approach is also a common path planning approach. The hybrid approach combines the traditional approach with the machine learning approach to improve the performance of path planning by comprehensively utilizing their points. For example, heuristic searching in conventional approaches may be combined with machine learning models to achieve more efficient path planning. There are also some path planning methods that are specific to dynamic pedestrian environments. For example, a perception model-based approach utilizes sensor data to detect and predict pedestrians, thereby accounting for the motion trend of pedestrians in path planning. In addition, some collaborative path planning methods realize safe interaction between the unmanned aerial vehicle and the pedestrian by communicating and collaborating with the pedestrian. However, path planning for unmanned aerial vehicles in a dynamic pedestrian environment still faces some challenges.
First, the motion and behavior of pedestrians have uncertainties, requiring an accurate pedestrian prediction model. Secondly, path planning needs to consider the balance of a plurality of indexes such as path length, time efficiency, safety and the like. In addition, real-time performance and computational complexity are also factors to be considered in the path planning method.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for planning a meal delivery path of an unmanned aerial vehicle, which are used for solving the technical problems that the conventional unmanned aerial vehicle is poor in meal delivery path planning and easy to collide.
The invention provides a meal delivery path planning method of an unmanned aerial vehicle, which comprises the following steps:
Acquiring scene information of an unmanned aerial vehicle flight environment, and acquiring pedestrian position information from the scene information;
Generating a pedestrian prediction track according to the pedestrian position information;
Generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track;
optimizing the initial flight path to obtain a target flight path;
and planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters.
Optionally, the step of collecting scene information of the unmanned aerial vehicle flight environment and acquiring pedestrian position information from the scene information includes:
collecting a depth image of the unmanned aerial vehicle flight environment;
converting the depth image into point cloud data;
Converting the point cloud data into a grid map;
Pedestrian detection is carried out on the depth image, so that pedestrian information is obtained;
And matching the pedestrian information in the raster image to obtain pedestrian position information corresponding to the pedestrian information.
Optionally, the step of generating a pedestrian prediction track according to the pedestrian position information includes:
Based on the current moment, acquiring a plurality of position points in the pedestrian position information at preset time intervals, and generating an actual path point set;
acquiring position coordinates of all position points in the actual path point set;
Calculating the speed of the pedestrian between two adjacent position points by adopting the position coordinates and the preset time interval;
calculating the pedestrian acceleration by adopting the pedestrian speed;
calculating a predicted displacement using the pedestrian speed and the pedestrian acceleration;
calculating a track curvature value by adopting position coordinates of every three adjacent position points;
calculating the curvature change speed by adopting the track curvature value;
calculating curvature change acceleration by adopting the curvature change speed;
calculating the curvature of the predicted point by adopting the track curvature value, the curvature change speed and the curvature change acceleration;
Determining a predicted path point by adopting the position coordinates of the position points, the predicted displacement and the curvature of the predicted point;
Judging whether the current prediction times are smaller than preset prediction times or not;
If yes, adding the predicted path point into the actual path point set, and returning to the step of acquiring the position coordinates of all the position points in the actual path point set;
and if not, generating a pedestrian prediction track by adopting the actual path point set.
Optionally, the step of generating the initial flight path of the unmanned aerial vehicle according to the pedestrian predicted trajectory includes:
Calculating pedestrian collision cost of the unmanned aerial vehicle on each grid of the grid map according to the pedestrian prediction track;
calculating the current cost and the estimated cost of the unmanned aerial vehicle in each grid;
Generating node cost of the unmanned aerial vehicle in each grid by adopting the current cost, the estimated cost and the pedestrian collision cost;
and generating an initial flight path of the unmanned aerial vehicle according to the node cost.
Optionally, the flight parameters include a speed, a direction, and a pose of the drone.
The invention also provides an unmanned aerial vehicle meal delivery path planning device, which comprises:
the pedestrian position information acquisition module is used for acquiring scene information of the unmanned aerial vehicle flight environment and acquiring pedestrian position information from the scene information;
the pedestrian prediction track generation module is used for generating a pedestrian prediction track according to the pedestrian position information;
The initial flight path generation module is used for generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track;
The target flight path acquisition module is used for optimizing the initial flight path to obtain a target flight path;
and the meal delivery task execution module is used for planning flight parameters according to the target flight path and executing meal delivery tasks according to the flight parameters.
Optionally, the pedestrian position information acquisition module includes:
the depth image acquisition sub-module is used for acquiring a depth image of the unmanned aerial vehicle flight environment;
the point cloud data conversion sub-module is used for converting the depth image into point cloud data;
The grid map conversion sub-module is used for converting the point cloud data into a grid map;
the pedestrian detection sub-module is used for detecting pedestrians on the depth image to obtain pedestrian information;
And the pedestrian position information acquisition sub-module is used for matching the pedestrian information in the raster image to obtain pedestrian position information corresponding to the pedestrian information.
Optionally, the pedestrian prediction track generation module includes:
The actual path point set generation sub-module is used for acquiring a plurality of position points in the pedestrian position information at preset time intervals based on the current moment to generate an actual path point set;
the position coordinate acquisition sub-module is used for acquiring the position coordinates of all the position points in the actual path point set;
The pedestrian speed calculation sub-module is used for calculating the pedestrian speed between two adjacent position points by adopting the position coordinates and the preset time interval;
the pedestrian acceleration calculation sub-module is used for calculating the pedestrian acceleration by adopting the pedestrian speed;
a predicted displacement calculation sub-module for calculating a predicted displacement using the pedestrian speed and the pedestrian acceleration;
the track curvature value calculation sub-module is used for calculating a track curvature value by adopting the position coordinates of every three adjacent position points;
a curvature change speed calculation sub-module for calculating a curvature change speed by using the track curvature value;
A curvature change acceleration calculation sub-module for calculating curvature change acceleration using the curvature change speed;
The predicted point curvature calculation submodule is used for calculating the predicted point curvature by adopting the track curvature value, the curvature change speed and the curvature change acceleration;
A predicted path point determination sub-module for determining a predicted path point using the position coordinates of the position point, the predicted displacement, and the predicted point curvature;
The judging sub-module is used for judging whether the current predicted times are smaller than preset predicted times or not;
a return sub-module, configured to, if yes, add the predicted path point into the actual path point set, and return to the step of obtaining position coordinates of all the position points in the actual path point set;
and the pedestrian prediction track generation sub-module is used for generating a pedestrian prediction track by adopting the actual path point set if not.
The invention also provides an electronic device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
The processor is configured to execute the unmanned aerial vehicle meal delivery path planning method according to any one of the above instructions in the program code.
The invention also provides a computer readable storage medium for storing program code for performing the unmanned aerial vehicle meal delivery path planning method as defined in any one of the above.
From the above technical scheme, the invention has the following advantages: according to the invention, the scene information of the unmanned aerial vehicle flight environment is acquired, and the pedestrian position information is acquired from the scene information; generating a pedestrian prediction track according to the pedestrian position information; generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track; optimizing an initial flight path to obtain a target flight path; and planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters. The unmanned aerial vehicle can consider the running of pedestrians to carry out path planning in a dynamic pedestrian environment, so that the safety of the unmanned aerial vehicle flight task is effectively improved, and the occurrence of flight accidents is reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained from these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a step flowchart of an unmanned aerial vehicle meal delivery path planning method provided by an embodiment of the invention;
fig. 2 is a flowchart of steps of a method for planning a meal delivery path of an unmanned aerial vehicle according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of a matlab modeling pedestrian prediction trajectory prediction process;
FIG. 4 is a schematic diagram of a grid node and pedestrian motion relationship;
fig. 5 is a flow chart for calculating an optimal initial path by improving the a-algorithm;
FIG. 6 is a schematic diagram of a unmanned aerial vehicle trajectory tracking process;
fig. 7 is a block diagram of a configuration of an unmanned aerial vehicle meal delivery path planning device according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a method, a device, equipment and a storage medium for planning a meal delivery path of an unmanned aerial vehicle, which are used for solving the technical problems that the conventional unmanned aerial vehicle is poor in meal delivery path planning and easy to collide.
In order to make the objects, features and advantages of the present invention more comprehensible, the technical solutions in the embodiments of the present invention are described in detail below with reference to the accompanying drawings, and it is apparent that the embodiments described below are only some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, fig. 1 is a flowchart illustrating steps of a method for planning a meal delivery path of an unmanned aerial vehicle according to an embodiment of the present invention.
The invention provides a unmanned aerial vehicle meal delivery path planning method, which specifically comprises the following steps:
step 101, acquiring scene information of an unmanned aerial vehicle flight environment, and acquiring pedestrian position information from the scene information;
Unmanned aerial vehicles, which are unmanned aerial vehicles that are operated by means of a radio remote control and a self-contained program control, or are operated autonomously, either entirely or intermittently, by an on-board computer.
In the embodiment of the invention, the unmanned aerial vehicle can acquire scene information of the unmanned aerial vehicle flight environment through the depth camera in the meal delivery process, and detect pedestrians from the scene information to obtain pedestrian position information.
102, Generating a pedestrian prediction track according to pedestrian position information;
According to the embodiment of the invention, the track change of the pedestrians in a certain time can be predicted according to the acquired pedestrian position information of the pedestrians at different moments, the possible track information of the pedestrians in the unmanned aerial vehicle flight process is obtained, and the pedestrian prediction track of the pedestrians is generated.
Step 103, generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track;
In the embodiment of the invention, after the predicted track of the pedestrian in the scene in a certain time is predicted, the predicted track of the pedestrian can be used as an obstacle to generate the initial flight path of the unmanned aerial vehicle.
104, Optimizing an initial flight path to obtain a target flight path;
After the initial flight path of the unmanned aerial vehicle is generated, the unmanned aerial vehicle can acquire scene information in real time in the flight process to optimize the initial flight path, so that a target flight path is obtained.
And 105, planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters.
After the target flight path of the unmanned aerial vehicle is obtained, the flight parameters of the unmanned aerial vehicle can be planned according to the target flight path, and the flight state of the unmanned aerial vehicle can be adjusted according to the flight parameters, so that the execution of the meal delivery task is completed.
According to the invention, the scene information of the unmanned aerial vehicle flight environment is acquired, and the pedestrian position information is acquired from the scene information; generating a pedestrian prediction track according to the pedestrian position information; generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track; optimizing an initial flight path to obtain a target flight path; and planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters. The unmanned aerial vehicle can consider the running of pedestrians to carry out path planning in a dynamic pedestrian environment, so that the safety of the unmanned aerial vehicle flight task is effectively improved, and the occurrence of flight accidents is reduced.
Referring to fig. 2, fig. 2 is a flowchart illustrating steps of a method for planning a meal delivery path of an unmanned aerial vehicle according to another embodiment of the present invention. The method specifically comprises the following steps:
step 201, collecting a depth image of the unmanned aerial vehicle flight environment;
step 202, converting the depth image into point cloud data;
step 203, converting the point cloud data into a grid map;
Step 204, pedestrian detection is carried out on the depth image, and pedestrian information is obtained;
step 205, matching pedestrian information in the raster image to obtain pedestrian position information corresponding to the pedestrian information;
a depth image is a grayscale image in which each pixel point represents the distance of each corresponding point from the camera.
And (3) point cloud: the object is represented by dots, and the formed cluster of dots is called a point cloud like a cloud. The point cloud data represents coordinate information of three-dimensional points in the scene. The coordinates of each point may be calculated from the pixel coordinates in the depth image and the depth value of the corresponding pixel.
A grid map is a three-dimensional grid structure that represents obstacles, spatial information, and other attributes in the environment.
In a particular implementation, a depth camera may be used to capture a depth image in a scene of an unmanned aerial vehicle flight environment, and then the depth image is converted to point cloud data. After the point cloud data is obtained, filtering processing can be performed on the point cloud data to remove noise and invalid points. Common filtering methods include kalman filtering, gaussian filtering, statistical filtering, voxel filtering and the like, and a person skilled in the art can select a corresponding filtering mode according to actual requirements, which is not particularly limited in the embodiment of the present invention.
After obtaining the point cloud data, the point cloud data may be converted into a grid map, specifically, each point in the point cloud may be mapped into a corresponding grid cell, and the values of the grid cells may be set as needed, for example, by representing an obstacle or a feasible region or the like with different values.
After the grid map of the unmanned aerial vehicle flight environment scene is obtained, pedestrian detection can be carried out on the preprocessed depth image or point cloud data by using a YOLO pedestrian detection algorithm so as to obtain pedestrian information, and then corresponding pedestrian information is matched in the grid map so as to obtain the pedestrian position information of each pedestrian.
The YOLO is a real-time target detection algorithm based on deep learning, and can rapidly and accurately detect pedestrians in images or scenes and acquire position information of the pedestrians.
Step 206, generating a pedestrian prediction track according to the pedestrian position information;
After the pedestrian position information is acquired, the pedestrian prediction track of the pedestrian in a certain time can be predicted according to the pedestrian position information.
In one example, the step of generating a pedestrian prediction track from pedestrian location information may include the sub-steps of:
S601, acquiring a plurality of position points in pedestrian position information at preset time intervals based on the current moment, and generating an actual path point set;
s602, obtaining position coordinates of all position points in an actual path point set;
S603, calculating the speed of the pedestrian between two adjacent position points by adopting the position coordinates and a preset time interval;
S604, calculating the acceleration of the pedestrian by adopting the speed of the pedestrian;
s605, calculating predicted displacement by using the pedestrian speed and the pedestrian acceleration;
S606, calculating a track curvature value by adopting position coordinates of every three adjacent position points;
s607, calculating the curvature change speed by adopting the track curvature value;
s608, calculating curvature change acceleration by using the curvature change speed;
s609, calculating the curvature of the predicted point by adopting the track curvature value, the curvature change speed and the curvature change acceleration;
S610, determining a predicted path point by using the position coordinates of the position points, the predicted displacement and the curvature of the predicted point;
s611, judging whether the current prediction times are smaller than preset prediction times or not;
S612, if yes, adding the predicted path point into the actual path point set, and returning to the step of acquiring the position coordinates of all the position points in the actual path point set;
s613, if not, generating a pedestrian prediction track by adopting the actual path point set.
In a specific implementation, the prediction of the predicted track of the pedestrian can firstly obtain the position points of the pedestrian at the current moment and a plurality of moments before the current moment in the acquired pedestrian position information, and generate an actual path point set. If the current time and the four position points at the previous time are recorded in the actual path point set P, the time interval between the five position points meeting the adjacent two points is Δt. And then predicting the pedestrian prediction track according to the recorded actual path point set, wherein the process is as follows:
Firstly, initializing an actual path point set of a pedestrian prediction track, setting a time interval delta t and setting the number of prediction points as n, taking the motion speed and the motion curvature of a pedestrian into consideration in the pedestrian prediction track prediction method, and assuming that the last 5 position points of the pedestrian prediction track are P t-4,Pt-3,Pt-2,Pt-1,Pt, wherein the path point P * represents two-dimensional vector coordinates P t is denoted as the pedestrian current time position.
The pedestrian speed is estimated as:
the pedestrian acceleration is estimated as:
where Δt represents the time interval of two adjacent waypoints, whereby the predicted displacement Δl of the pedestrian at the next Δt time is estimated as:
Then, curvature estimation is carried out on pedestrian motion, if three continuous path points are collinear, the curvature is 0, if the three path points are not collinear, a circle can be determined, the radius of the circle is r, and the curvature curve is calculated as: The curvature cut at this time is constant as a positive number. The curvature of the prescribed point P t is determined by P t-2,Pt-1,Pt, and considering the convexity of the curve, the curvature is positive when the prescribed curve is concave, the curvature is negative when the curve is convex, and the convexity of the curve is determined by the y coordinates:
From 5 path points, 3 curvature values are calculated: the curvature cur t at point P t is determined by P t-2,Pt-1,Pt, the curvature cur t-1 at point P t-1 is determined by P t-3,Pt-2,Pt-1, and the curvature cur t-2 at point P t-2 is determined by P t-4,Pt-3,Pt-2.
The curvature change speed calculation estimation of the pedestrian predicted trajectory is as follows:
wherein cur.v * represents the curvature change speed at the moment, and the curvature change acceleration calculation estimation of the pedestrian predicted trajectory is:
Where cur.a * represents the curvature change acceleration at time. From the curvature change information calculated above, a predicted point curvature cur t+1 is calculated:
The predicted motion displacement and the predicted curvature change of the pedestrian are obtained through the calculation, and the next track point of the pedestrian is finally predicted by combining the last two points P t,Pt-1 in the track point set. And obtaining a circle O by using the P t,Pt-1 and the predicted curvature cur t+1, simulating the predicted track motion of the pedestrian as P t-1-＞Pt-＞Pt+1, obtaining a predicted path point P t+1 by using the distance of the P t moving along the motion direction in the circle O, adding the predicted point P t+1 into the path point set, and completing the prediction of one path point. And (3) according to the number n of the initialized predicted points, iteratively calculating n to finish n predicted points to obtain a predicted path point set waypoints _pre.
After the predicted path point set is obtained, a quasi-uniform B-spline curve can be used to represent the predicted pedestrian trajectory according to the predicted path point set waypoints _pre. Assuming a total of n+1 path points { P 0,P1,...,Pn }, which are used to define the trend, bounding range of the B-spline curve, the definition of the k-order B-spline curve is:
wherein B i,k (u) is the ith k-th order B-spline basis function, corresponding to the path point P i, k.gtoreq.1, u being an argument. The basis functions have the following debulk-koxz recursion:
Convention 0/0=0, where u i is a set of continuously varying values called a non-decreasing sequence of node vectors, the sequence being (u 0,u1,...,uk,uk+1,...,un,un+1,...,un+k), with a cubic quasi-uniform B-spline curve, k=3, and the node vectors being (0,0,0,0,1,2, n-4, n-3). Fig. 3 is a schematic diagram of a matlab process for simulating prediction of a pedestrian prediction track, wherein the first figure is an actual motion path point of a pedestrian, the last three figures are the results of the iterative prediction of fig. 3, and the blocks are prediction parts.
Step 207, generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track;
In the embodiment of the invention, after the predicted track of the pedestrian in the scene in a certain time is predicted, the predicted track of the pedestrian can be used as an obstacle to generate the initial flight path of the unmanned aerial vehicle.
In one example, the step of generating an initial flight path of the drone from the pedestrian predicted trajectory may include the sub-steps of:
s71, calculating pedestrian collision cost of the unmanned aerial vehicle on each grid of the grid map according to the pedestrian prediction track;
S72, calculating the current cost and the estimated cost of the unmanned aerial vehicle in each grid;
s73, generating node cost of the unmanned aerial vehicle in each grid by adopting the current cost, the estimated cost and the pedestrian collision cost;
s74, generating an initial flight path of the unmanned aerial vehicle according to the node cost.
In a specific implementation, an initial path can be generated according to the starting point and the target point of the unmanned aerial vehicle and the prediction result of the pedestrian prediction track so as to adapt to the environment and task requirements. The pedestrian prediction track part can be regarded as a dangerous area, the pedestrian prediction track part can be treated as an obstacle point in path planning, and the selection of the flight path point can adopt an optimization A-type algorithm. The improved cost function of the optimization a-algorithm is as follows:
f(n)＝g(n)+h(n)+behind(n)
Where f (n) is the node cost and g (n) is the current cost, i.e. how many steps the starting point has taken to the current grid.
H (n) is used as the estimated cost, which indicates how many steps are required to be taken from the current grid to the destination grid, the estimated cost can be represented by Euclidean distance, namely, the linear distance between two points, and the formula in the two-dimensional space is calculated as follows:
The behend (n) is a pedestrian collision cost, and represents a cost value near the pedestrian, if the grid point is positioned at the front of the pedestrian movement direction, a higher collision cost is generated, and if the grid point is positioned at the rear of the pedestrian movement direction, the collision cost is lower, so that the collision cost is weighted for the path points near the pedestrian. Weight calculation according to the schematic diagram of the relation between the grid node and the motion of the pedestrian in fig. 4, when the distance between the unmanned aerial vehicle and the pedestrian is greater than the distance threshold D, the collision cost behend (n) =0, and when the distance between the unmanned aerial vehicle and the pedestrian is less than the distance threshold D (the distance threshold D is a value set in advance), the collision cost behend (n) of the pedestrian is calculated according to a weight formula:
Wherein n represents the position of the search node, and α is used for judging whether the distance between the unmanned aerial vehicle and the pedestrian is smaller than a distance threshold D:
Beta is used for judging the relation between the search node n and the pedestrian movement direction, the value of beta 1,β2 needs to be adjusted, and beta 1＜β2 is as follows:
and then, according to the improved cost function, an improved A-algorithm is realized to find the optimal initial path from the starting point to the end point, and the optimal initial path is used as the initial flight path of the unmanned aerial vehicle.
As shown in fig. 5, fig. 5 is a flowchart for calculating an optimal initial path by modifying the a algorithm.
1. Adding the starting point of the unmanned aerial vehicle into an open list;
2. traversing open list, searching the node with the minimum f (n), taking the node as the node to be processed currently, and adding the node into close list;
3. searching adjacent nodes of the current node;
4. judging whether the searched adjacent node is an obstacle or is in a close list, if so, returning to the step 3; if not, entering a step 5;
5. Judging whether the searched adjacent node is in an open list, if not, adding the adjacent node into the open list, setting the current grid as a father node, recording the values of f (n), g (n), h (n) and behend (n) of the grid, and returning to the step 3; if yes, checking whether the path is better or not, and updating the father node;
6. Judging whether the grid map is searched, if not, returning to the step 3; if yes, judging whether the end point coordinates are in an open list; if the terminal point coordinates are not in the open list, judging that an initial flight path of the unmanned aerial vehicle does not exist; and if the end point coordinates exist in the open list, taking the current path as an initial flight path of the unmanned aerial vehicle.
Step 208, optimizing the initial flight path to obtain a target flight path;
after the initial flight path is obtained, the initial flight path may be optimized to meet specific goals and constraints. The optimization algorithm may adjust and refine the path based on different optimization objectives, such as shortest path, minimum energy consumption, or minimum time of flight.
In a specific implementation, assuming that N path points q 0,q1,......,qN-1 exist in the obtained initial flight path, the whole track is divided into N-1 sections of independent track combinations, and the time allocation of each section of track is initialized toWherein v max is the maximum flight speed at which the drone is flown. The trace is fitted into a polynomial form by adopting a method of trace fitting of a minimum snap polynomial: p (t) =p 0t0+p1t1+...+pntn, n=7. The snap is the second derivative of acceleration, and correspondingly the inverse of the thrust of the unmanned aerial vehicle, and the minimization of the snap enables the thrust change speed to be minimized, so that energy is saved. The cost function defined by each segment of track is: /(I)Adding the cost of each section of track, and adding the boundary point condition and the intermediate point condition to obtain the cost J=J 1+J2+......+JN-1 of the whole track. Considering the continuity requirement of the whole track, the speed and acceleration of each path point must be the same, and a constraint of the track is formed. And placing the initial conditions into a QP solver to obtain an optimal polynomial parameter matrix, generating an optimal track of unmanned aerial vehicle flight, and taking the optimal track as a target flight path. The initial conditions include: the pose (position, speed, acceleration) of the start and end points of the unmanned aerial vehicle, the point of the path in flight, and the time allocated for each small track.
Step 209, planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters.
In the embodiment of the invention, the flight behavior of the unmanned aerial vehicle is controlled according to the finally determined target flight path. This includes controlling the speed, direction and attitude of the aircraft to achieve a planned path. And then controlling the unmanned aerial vehicle to execute the meal delivery task according to the target flight path. The method for completing the unmanned aerial vehicle track tracking process is as shown in fig. 6, and the unmanned aerial vehicle is controlled to fly according to the designed track. The classical MPC model predictive control is adopted, and is essentially feedback control, after a group of control outputs are obtained through an optimization method, the unmanned aerial vehicle executes control instructions, and the state z t of the unmanned aerial vehicle of the current task is continuously fed back at a certain frequency. The state is input to the path planning module and the MPC control module at the same time, and the path planning module makes planning again according to the new unmanned plane state and combining the information of the perception module and the map information. And the MPC performs a new round of predictive control according to the new reference path and the current state of the unmanned aerial vehicle.
According to the invention, the scene information of the unmanned aerial vehicle flight environment is acquired, and the pedestrian position information is acquired from the scene information; generating a pedestrian prediction track according to the pedestrian position information; generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track; optimizing an initial flight path to obtain a target flight path; and planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters. The unmanned aerial vehicle can consider the running of pedestrians to carry out path planning in a dynamic pedestrian environment, so that the safety of the unmanned aerial vehicle flight task is effectively improved, and the occurrence of flight accidents is reduced.
Referring to fig. 7, fig. 7 is a block diagram illustrating a configuration of an unmanned aerial vehicle meal delivery path planning apparatus according to an embodiment of the present invention.
The embodiment of the invention provides an unmanned aerial vehicle meal delivery path planning device, which comprises the following steps:
The pedestrian position information acquisition module 701 is configured to acquire scene information of a flight environment of the unmanned aerial vehicle, and acquire pedestrian position information from the scene information;
A pedestrian prediction track generation module 702, configured to generate a pedestrian prediction track according to pedestrian position information;
an initial flight path generation module 703, configured to generate an initial flight path of the unmanned aerial vehicle according to the pedestrian predicted trajectory;
A target flight path acquisition module 704, configured to optimize an initial flight path to obtain a target flight path;
The meal delivery task execution module 705 is configured to plan flight parameters according to the target flight path, and execute a meal delivery task according to the flight parameters.
In the embodiment of the present invention, the pedestrian position information acquisition module 701 includes:
the depth image acquisition sub-module is used for acquiring a depth image of the unmanned aerial vehicle flight environment;
The point cloud data conversion sub-module is used for converting the depth image into point cloud data;
the grid map conversion sub-module is used for converting the point cloud data into a grid map;
the pedestrian detection sub-module is used for detecting pedestrians on the depth image to obtain pedestrian information;
And the pedestrian position information acquisition sub-module is used for matching the pedestrian information in the grid image to obtain pedestrian position information corresponding to the pedestrian information.
In an embodiment of the present invention, the pedestrian prediction track generation module 702 includes:
The actual path point set generation sub-module is used for acquiring a plurality of position points in the pedestrian position information at preset time intervals based on the current moment to generate an actual path point set;
The position coordinate acquisition sub-module is used for acquiring the position coordinates of all the position points in the actual path point set;
The pedestrian speed calculation sub-module is used for calculating the pedestrian speed between two adjacent position points by adopting the position coordinates and a preset time interval;
The pedestrian acceleration calculation sub-module is used for calculating the pedestrian acceleration by adopting the pedestrian speed;
the predicted displacement calculation sub-module is used for calculating predicted displacement by adopting the pedestrian speed and the pedestrian acceleration;
The track curvature value calculation sub-module is used for calculating a track curvature value by adopting the position coordinates of every three adjacent position points;
The curvature change speed calculation submodule is used for calculating the curvature change speed by adopting the track curvature value;
The curvature change acceleration calculation sub-module is used for calculating curvature change acceleration by adopting the curvature change speed;
the predicted point curvature calculation submodule is used for calculating the predicted point curvature by adopting a track curvature value, a curvature change speed and a curvature change acceleration;
The predicted path point determination submodule is used for determining a predicted path point by adopting position coordinates of the position point, predicted displacement and curvature of the predicted point;
The judging sub-module is used for judging whether the current predicted times are smaller than preset predicted times or not;
the return sub-module is used for adding the predicted path point into the actual path point set if yes, and returning to the step of acquiring the position coordinates of all the position points in the actual path point set;
And the pedestrian prediction track generation sub-module is used for generating a pedestrian prediction track by adopting the actual path point set if not.
In an embodiment of the present invention, the initial flight path generation module 703 includes:
The pedestrian collision cost calculation sub-module is used for calculating the pedestrian collision cost of the unmanned aerial vehicle on each grid of the grid map according to the pedestrian prediction track;
The current cost and estimated cost calculation sub-module is used for calculating the current cost and estimated cost of the unmanned aerial vehicle in each grid;
The node cost calculation sub-module is used for generating the node cost of the unmanned aerial vehicle in each grid by adopting the current cost, the estimated cost and the pedestrian collision cost;
and the initial flight path generation sub-module is used for generating an initial flight path of the unmanned aerial vehicle according to the node cost.
In an embodiment of the invention, the flight parameters include speed, direction and attitude of the unmanned aerial vehicle.
The embodiment of the invention also provides electronic equipment, which comprises a processor and a memory:
The memory is used for storing the program codes and transmitting the program codes to the processor;
the processor is used for executing the unmanned aerial vehicle meal delivery path planning method according to the instructions in the program codes.
The embodiment of the invention also provides a computer readable storage medium, which is used for storing program codes, and the program codes are used for executing the unmanned aerial vehicle meal delivery path planning method.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or terminal device that comprises the element.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (5)
1. The unmanned aerial vehicle meal delivery path planning method is characterized by comprising the following steps of:
Acquiring scene information of an unmanned aerial vehicle flight environment, and acquiring pedestrian position information from the scene information;
Generating a pedestrian prediction track according to the pedestrian position information;
Generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track;
optimizing the initial flight path to obtain a target flight path;
Planning flight parameters according to the target flight path, and executing meal delivery tasks according to the flight parameters;
the step of acquiring scene information of the unmanned aerial vehicle flight environment and acquiring pedestrian position information from the scene information comprises the following steps:
collecting a depth image of the unmanned aerial vehicle flight environment;
converting the depth image into point cloud data;
Converting the point cloud data into a grid map;
Pedestrian detection is carried out on the depth image, so that pedestrian information is obtained;
Matching the pedestrian information in the grid map to obtain pedestrian position information corresponding to the pedestrian information;
The step of generating a pedestrian prediction track according to the pedestrian position information comprises the following steps:
Based on the current moment, acquiring a plurality of position points in the pedestrian position information at preset time intervals, and generating an actual path point set;
acquiring position coordinates of all position points in the actual path point set;
Calculating the speed of the pedestrian between two adjacent position points by adopting the position coordinates and the preset time interval;
calculating the pedestrian acceleration by adopting the pedestrian speed;
calculating a predicted displacement using the pedestrian speed and the pedestrian acceleration;
calculating a track curvature value by adopting position coordinates of every three adjacent position points;
calculating the curvature change speed by adopting the track curvature value;
calculating curvature change acceleration by adopting the curvature change speed;
calculating the curvature of the predicted point by adopting the track curvature value, the curvature change speed and the curvature change acceleration;
Determining a predicted path point by adopting the position coordinates of the position points, the predicted displacement and the curvature of the predicted point;
Judging whether the current prediction times are smaller than preset prediction times or not;
If yes, adding the predicted path point into the actual path point set, and returning to the step of acquiring the position coordinates of all the position points in the actual path point set;
If not, generating a pedestrian prediction track by adopting the actual path point set;
The step of generating the initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track comprises the following steps:
Calculating pedestrian collision cost of the unmanned aerial vehicle on each grid of the grid map according to the pedestrian prediction track;
calculating the current cost and the estimated cost of the unmanned aerial vehicle in each grid;
Generating node cost of the unmanned aerial vehicle in each grid by adopting the current cost, the estimated cost and the pedestrian collision cost;
and generating an initial flight path of the unmanned aerial vehicle according to the node cost.
2. The method of claim 1, wherein the flight parameters include speed, direction, and attitude of the drone.
3. Unmanned aerial vehicle meal delivery path planning device, its characterized in that includes:
the pedestrian position information acquisition module is used for acquiring scene information of the unmanned aerial vehicle flight environment and acquiring pedestrian position information from the scene information;
the pedestrian prediction track generation module is used for generating a pedestrian prediction track according to the pedestrian position information;
The initial flight path generation module is used for generating an initial flight path of the unmanned aerial vehicle according to the pedestrian prediction track;
The target flight path acquisition module is used for optimizing the initial flight path to obtain a target flight path;
The meal delivery task execution module is used for planning flight parameters according to the target flight path and executing meal delivery tasks according to the flight parameters;
Wherein, pedestrian position information acquisition module includes:
the depth image acquisition sub-module is used for acquiring a depth image of the unmanned aerial vehicle flight environment;
the point cloud data conversion sub-module is used for converting the depth image into point cloud data;
The grid map conversion sub-module is used for converting the point cloud data into a grid map;
the pedestrian detection sub-module is used for detecting pedestrians on the depth image to obtain pedestrian information;
the pedestrian position information acquisition sub-module is used for matching the pedestrian information in the grid map to obtain pedestrian position information corresponding to the pedestrian information;
the pedestrian prediction track generation module comprises:
The actual path point set generation sub-module is used for acquiring a plurality of position points in the pedestrian position information at preset time intervals based on the current moment to generate an actual path point set;
the position coordinate acquisition sub-module is used for acquiring the position coordinates of all the position points in the actual path point set;
The pedestrian speed calculation sub-module is used for calculating the pedestrian speed between two adjacent position points by adopting the position coordinates and the preset time interval;
the pedestrian acceleration calculation sub-module is used for calculating the pedestrian acceleration by adopting the pedestrian speed;
a predicted displacement calculation sub-module for calculating a predicted displacement using the pedestrian speed and the pedestrian acceleration;
the track curvature value calculation sub-module is used for calculating a track curvature value by adopting the position coordinates of every three adjacent position points;
a curvature change speed calculation sub-module for calculating a curvature change speed by using the track curvature value;
A curvature change acceleration calculation sub-module for calculating curvature change acceleration using the curvature change speed;
The predicted point curvature calculation submodule is used for calculating the predicted point curvature by adopting the track curvature value, the curvature change speed and the curvature change acceleration;
A predicted path point determination sub-module for determining a predicted path point using the position coordinates of the position point, the predicted displacement, and the predicted point curvature;
The judging sub-module is used for judging whether the current predicted times are smaller than preset predicted times or not;
a return sub-module, configured to, if yes, add the predicted path point into the actual path point set, and return to the step of obtaining position coordinates of all the position points in the actual path point set;
The pedestrian prediction track generation sub-module is used for generating a pedestrian prediction track by adopting the actual path point set if not;
Wherein, initial flight path generates module includes:
The pedestrian collision cost calculation sub-module is used for calculating the pedestrian collision cost of the unmanned aerial vehicle on each grid of the grid map according to the pedestrian prediction track;
The current cost and estimated cost calculation sub-module is used for calculating the current cost and estimated cost of the unmanned aerial vehicle in each grid;
The node cost calculation sub-module is used for generating the node cost of the unmanned aerial vehicle in each grid by adopting the current cost, the estimated cost and the pedestrian collision cost;
and the initial flight path generation sub-module is used for generating an initial flight path of the unmanned aerial vehicle according to the node cost.
4. An electronic device, the device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the unmanned aerial vehicle meal delivery path planning method according to any one of claims 1-2 according to instructions in the program code.
5. A computer readable storage medium for storing program code for performing the unmanned aerial vehicle meal delivery path planning method of any one of claims 1-2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311136458.5A CN117055601B (en) | 2023-09-05 | 2023-09-05 | Unmanned aerial vehicle meal delivery path planning method, unmanned aerial vehicle meal delivery path planning device, unmanned aerial vehicle meal delivery path planning equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311136458.5A CN117055601B (en) | 2023-09-05 | 2023-09-05 | Unmanned aerial vehicle meal delivery path planning method, unmanned aerial vehicle meal delivery path planning device, unmanned aerial vehicle meal delivery path planning equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117055601A CN117055601A (en) | 2023-11-14 |
| CN117055601B true CN117055601B (en) | 2024-05-31 |
Family
ID=88653544
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311136458.5A Active CN117055601B (en) | 2023-09-05 | 2023-09-05 | Unmanned aerial vehicle meal delivery path planning method, unmanned aerial vehicle meal delivery path planning device, unmanned aerial vehicle meal delivery path planning equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117055601B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108776488A (en) * | 2018-03-12 | 2018-11-09 | 徐晨旭 | A kind of method of path planning |
| CN109116867A (en) * | 2018-09-28 | 2019-01-01 | 拓攻（南京）机器人有限公司 | A kind of unmanned plane during flying barrier-avoiding method, device, electronic equipment and storage medium |
| CN110687923A (en) * | 2019-11-08 | 2020-01-14 | 深圳市道通智能航空技术有限公司 | Unmanned aerial vehicle long-distance tracking flight method, device, equipment and storage medium |
| CN110873565A (en) * | 2019-11-21 | 2020-03-10 | 北京航空航天大学 | Unmanned aerial vehicle real-time path planning method for urban scene reconstruction |
| CN112506222A (en) * | 2020-12-10 | 2021-03-16 | 中国南方电网有限责任公司超高压输电公司天生桥局 | Unmanned aerial vehicle intelligent obstacle avoidance method and device |
| CN113568435A (en) * | 2021-09-24 | 2021-10-29 | 深圳火眼智能有限公司 | Unmanned aerial vehicle autonomous flight situation perception trend based analysis method and system |
| CN114625162A (en) * | 2022-02-10 | 2022-06-14 | 广东工业大学 | Hybrid algorithm-based optimal path planning method, system and medium for unmanned aerial vehicle |
| CN114677588A (en) * | 2022-03-18 | 2022-06-28 | 深圳市普渡科技有限公司 | Obstacle detection method, obstacle detection device, robot and storage medium |
| CN115016546A (en) * | 2022-08-10 | 2022-09-06 | 中国科学院自动化研究所 | Unmanned aerial vehicle three-dimensional path planning method and device, electronic equipment and storage medium |
| CN115686052A (en) * | 2022-11-22 | 2023-02-03 | 深圳市边界智控科技有限公司 | Unmanned aerial vehicle obstacle avoidance path planning method and device, computer equipment and storage medium |
| CN116300913A (en) * | 2023-03-03 | 2023-06-23 | 大连海事大学 | Unmanned ship multi-constraint local path planning method based on visual information |
| CN116627168A (en) * | 2023-05-23 | 2023-08-22 | 南京信息工程大学 | DWA algorithm-based power line inspection unmanned aerial vehicle path planning method, storage medium and device |
-
2023
- 2023-09-05 CN CN202311136458.5A patent/CN117055601B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108776488A (en) * | 2018-03-12 | 2018-11-09 | 徐晨旭 | A kind of method of path planning |
| CN109116867A (en) * | 2018-09-28 | 2019-01-01 | 拓攻（南京）机器人有限公司 | A kind of unmanned plane during flying barrier-avoiding method, device, electronic equipment and storage medium |
| CN110687923A (en) * | 2019-11-08 | 2020-01-14 | 深圳市道通智能航空技术有限公司 | Unmanned aerial vehicle long-distance tracking flight method, device, equipment and storage medium |
| CN110873565A (en) * | 2019-11-21 | 2020-03-10 | 北京航空航天大学 | Unmanned aerial vehicle real-time path planning method for urban scene reconstruction |
| CN112506222A (en) * | 2020-12-10 | 2021-03-16 | 中国南方电网有限责任公司超高压输电公司天生桥局 | Unmanned aerial vehicle intelligent obstacle avoidance method and device |
| CN113568435A (en) * | 2021-09-24 | 2021-10-29 | 深圳火眼智能有限公司 | Unmanned aerial vehicle autonomous flight situation perception trend based analysis method and system |
| CN114625162A (en) * | 2022-02-10 | 2022-06-14 | 广东工业大学 | Hybrid algorithm-based optimal path planning method, system and medium for unmanned aerial vehicle |
| CN114677588A (en) * | 2022-03-18 | 2022-06-28 | 深圳市普渡科技有限公司 | Obstacle detection method, obstacle detection device, robot and storage medium |
| CN115016546A (en) * | 2022-08-10 | 2022-09-06 | 中国科学院自动化研究所 | Unmanned aerial vehicle three-dimensional path planning method and device, electronic equipment and storage medium |
| CN115686052A (en) * | 2022-11-22 | 2023-02-03 | 深圳市边界智控科技有限公司 | Unmanned aerial vehicle obstacle avoidance path planning method and device, computer equipment and storage medium |
| CN116300913A (en) * | 2023-03-03 | 2023-06-23 | 大连海事大学 | Unmanned ship multi-constraint local path planning method based on visual information |
| CN116627168A (en) * | 2023-05-23 | 2023-08-22 | 南京信息工程大学 | DWA algorithm-based power line inspection unmanned aerial vehicle path planning method, storage medium and device |
Non-Patent Citations (2)
| Title |
|---|
| 基于深度强化学习的海战场目标搜寻路径规划;杨清清等;《***工程与电子技术》;20221130;第44卷(第11期);第3486-3495页 * |
| 基于逆向强化学习的无人机路径规划;杨秀霞等;《电光与控制》;20230831;第30卷(第8期);第1-7页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117055601A (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11567502B2 (en) | Autonomous exploration framework for indoor mobile robotics using reduced approximated generalized Voronoi graph | |
| Sathyamoorthy et al. | Densecavoid: Real-time navigation in dense crowds using anticipatory behaviors | |
| Drews et al. | Aggressive deep driving: Combining convolutional neural networks and model predictive control | |
| CN111578940B (en) | Indoor monocular navigation method and system based on cross-sensor transfer learning | |
| CN112888612A (en) | Autonomous vehicle planning | |
| Levine et al. | Information-rich path planning with general constraints using rapidly-exploring random trees | |
| Bipin et al. | Autonomous navigation of generic monocular quadcopter in natural environment | |
| Drews et al. | Aggressive deep driving: Model predictive control with a cnn cost model | |
| Leon et al. | A review of tracking, prediction and decision making methods for autonomous driving | |
| Tao et al. | SEER: Safe efficient exploration for aerial robots using learning to predict information gain | |
| Wang et al. | End-to-end interactive prediction and planning with optical flow distillation for autonomous driving | |
| Vinothkanna | A survey on novel estimation approach of motion controllers for self-driving cars | |
| Sonny et al. | Q-learning-based unmanned aerial vehicle path planning with dynamic obstacle avoidance | |
| Qin et al. | Deep imitation learning for autonomous navigation in dynamic pedestrian environments | |
| Ahmad et al. | End-to-end probabilistic depth perception and 3d obstacle avoidance using pomdp | |
| Akmandor et al. | Deep reinforcement learning based robot navigation in dynamic environments using occupancy values of motion primitives | |
| CN117055601B (en) | Unmanned aerial vehicle meal delivery path planning method, unmanned aerial vehicle meal delivery path planning device, unmanned aerial vehicle meal delivery path planning equipment and storage medium | |
| Dudarenko et al. | Robot navigation system in stochastic environment based on reinforcement learning on lidar data | |
| WO2023242223A1 (en) | Motion prediction for mobile agents | |
| Bayoumi et al. | Learning foresighted people following under occlusions | |
| Khalil et al. | Integration of motion prediction with end-to-end latent RL for self-driving vehicles | |
| Wang et al. | AGRNav: Efficient and Energy-Saving Autonomous Navigation for Air-Ground Robots in Occlusion-Prone Environments | |
| Bertoncini et al. | Fixed-Wing UAV Path Planning and Collision Avoidance using Nonlinear Model Predictive Control and Sensor-based Cloud Detection | |
| Alagić et al. | Design of mobile robot motion framework based on modified vector field histogram | |
| Petkovic et al. | Human motion prediction framework for safe flexible robotized warehouses |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |