CN117311369A - Multi-scene intelligent robot inspection method - Google Patents
Multi-scene intelligent robot inspection method Download PDFInfo
- Publication number
- CN117311369A CN117311369A CN202311606621.XA CN202311606621A CN117311369A CN 117311369 A CN117311369 A CN 117311369A CN 202311606621 A CN202311606621 A CN 202311606621A CN 117311369 A CN117311369 A CN 117311369A
- Authority
- CN
- China
- Prior art keywords
- robot
- path
- inspection
- data
- electric quantity
- 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.)
- Granted
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 154
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000002159 abnormal effect Effects 0.000 claims description 28
- 230000004044 response Effects 0.000 claims description 23
- 238000001514 detection method Methods 0.000 claims description 20
- 238000013507 mapping Methods 0.000 claims description 14
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000007726 management method Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 7
- 238000012795 verification Methods 0.000 claims description 6
- 230000001172 regenerating effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 6
- 238000007405 data analysis Methods 0.000 abstract description 4
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000005265 energy consumption Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000005094 computer simulation Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention relates to the technical field of data processing, and provides a multi-scene intelligent robot inspection method, which comprises the following steps: generating a robot inspection task and calling digital information of an inspection area; obtaining basic data interactively; outputting a path planning result and taking the path planning result as a standard reference path; performing path image acquisition; generating obstacle features and road surface features; configuring a pre-aiming distance and an obstacle avoidance reference; generating a regression path; the robot is controlled to finish the inspection task, the technical problems that the inspection data cannot be adapted to actual scene requirements, the quality of the obtained inspection data is low and the accuracy is insufficient are solved, the inspection efficiency and accuracy are greatly improved by configuring a pre-aiming point and generating a regression path according to the actual scene requirements, the inspection cost and risk are reduced, meanwhile, the robot can efficiently execute the inspection task through intelligent path planning and scene recognition, the quality and reliability of the inspection data are improved, and a powerful supporting technical effect is provided for subsequent data analysis and decision.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an intelligent inspection method for robots in multiple scenes.
Background
The intelligent inspection of the robots in multiple scenes can be widely applied to various fields, and the common inspection method of the robots mainly comprises the following steps: and (5) fixed path inspection and sensor auxiliary inspection. The fixed path inspection refers to that the robot performs inspection according to a preset path, is similar to a traditional fixed inspection line, and is applicable to inspection tasks with stable scene and clear path along a specified path, but is not flexible enough when encountering obstacles or path changes; the sensor auxiliary inspection is that the robot carries various sensor devices, such as a camera, an infrared sensor and the like, and realizes the functions of obstacle detection, distance measurement and the like by sensing the surrounding environment, and the sensor data is utilized to assist path planning and obstacle avoidance, so that inspection efficiency and accuracy are improved, but manual intervention is still needed to deal with the change of the complex environment when the complex environment is changed.
In summary, although the conventional robot inspection methods such as fixed path inspection and sensor-assisted inspection can improve inspection efficiency, the method is not flexible enough when encountering obstacles or path changes, so that a more intelligent robot inspection method is required to cope with more complex inspection tasks.
In summary, in the prior art, the robot inspection cannot adapt to the actual scene requirement, and the obtained inspection data has the technical problems of low quality and insufficient accuracy.
Disclosure of Invention
The application provides an intelligent inspection method for robots in multiple scenes, which aims at solving the technical problems that the inspection of the robots in the prior art cannot adapt to the actual scene requirement, and the quality of the obtained inspection data is low and the accuracy is not enough.
In view of the above problems, the application provides a multi-scenario intelligent robot inspection method.
In a first aspect of the disclosure, a multi-scenario intelligent robot inspection method is provided, where the method includes: generating a robot inspection task and calling digital information of an inspection area; the method comprises the steps of interactively obtaining basic data of the robot, wherein the basic data comprise size data, function data and electric quantity data; executing model initialization of a path planning model through the basic data, inputting the digital information and the inspection task into the path planning model, outputting a path planning result, and taking the path planning result as a standard reference path of the robot; controlling the robot to execute scene inspection based on the standard reference path, and executing path image acquisition of the standard reference path through an image acquisition device; performing obstacle and road surface recognition on the path image acquisition result to generate obstacle characteristics and road surface characteristics; configuring a pre-aiming distance and an obstacle avoidance reference according to the obstacle characteristics, the road surface characteristics and the basic data; generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path; and controlling the robot to finish the inspection task according to the regression path.
In another aspect of the disclosure, a multi-scenario robotic intelligent inspection system is provided, wherein the system comprises: the information calling module is used for generating a robot inspection task and calling digital information of an inspection area; the system comprises a basic data acquisition module, a control module and a control module, wherein the basic data acquisition module is used for interactively acquiring basic data of the robot, and the basic data comprises size data, function data and electric quantity data; the path planning module is used for executing model initialization of a path planning model through the basic data, inputting the digital information and the inspection task into the path planning model, outputting a path planning result, and taking the path planning result as a standard reference path of the robot; the scene inspection module is used for controlling the robot to execute scene inspection based on the standard reference path and executing path image acquisition of the standard reference path through the image acquisition device; the feature generation module is used for executing obstacle and road surface recognition on the path image acquisition result and generating obstacle features and road surface features; the path configuration module is used for configuring a pre-aiming distance and obstacle avoidance references according to the obstacle characteristics, the road surface characteristics and the basic data; the regression path generation module is used for generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path; and the inspection control module is used for controlling the robot to finish the inspection task according to the return path.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
the method has the advantages that the robot inspection task is generated, and the digital information of the inspection area is called; the method comprises the steps of interactively obtaining basic data of the robot, wherein the basic data comprise size data, function data and electric quantity data; executing model initialization of a path planning model through basic data, inputting digital information and a patrol task into the path planning model, outputting a path planning result, and taking the path planning result as a standard reference path of the robot; controlling the robot to execute scene inspection based on the standard reference path, and executing path image acquisition of the standard reference path through the image acquisition device; performing obstacle and road surface recognition on the path image acquisition result to generate obstacle characteristics and road surface characteristics; configuring a pre-aiming distance and obstacle avoidance references according to the obstacle characteristics, the road surface characteristics and the basic data; generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path; the robot is controlled to complete the inspection task according to the regression path, the actual scene requirement is compared, the pre-aiming point is configured, the regression path is generated, the inspection efficiency and accuracy are greatly improved, the inspection cost and risk are reduced, meanwhile, the robot can efficiently execute the inspection task through intelligent path planning and scene recognition, the quality and reliability of inspection data are improved, and a powerful supporting technical effect is provided for subsequent data analysis and decision.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Fig. 1 is a schematic diagram of a possible flow of a multi-scenario intelligent inspection method for a robot according to an embodiment of the present application.
Fig. 2 is a schematic flow diagram of a possible path planning in a multi-scenario intelligent robot inspection method according to an embodiment of the present application.
Fig. 3 is a schematic flow chart of a possible regression path generation in the intelligent inspection method of the robot with multiple scenes according to the embodiment of the present application.
Fig. 4 is a schematic diagram of a possible structure of a multi-scenario intelligent inspection system for a robot according to an embodiment of the present application.
Reference numerals illustrate: the system comprises an information calling module 100, a basic data obtaining module 200, a path planning module 300, a scene inspection module 400, a feature generation module 500, a path configuration module 600, a regression path generation module 700 and an inspection control module 800.
Detailed Description
The embodiment of the application provides a multi-scene intelligent robot inspection method, which solves the technical problems that the inspection of a robot cannot be suitable for actual scene demands, the quality of the obtained inspection data is low and the accuracy is insufficient, realizes the comparison of the actual scene demands, configures a pre-aiming point, generates a regression path, greatly improves the inspection efficiency and accuracy, reduces the inspection cost and risk, and simultaneously, through intelligent path planning and scene recognition, the robot can efficiently execute the inspection task, improves the quality and reliability of the inspection data and provides a powerful supporting technical effect for subsequent data analysis and decision.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides a multi-scenario intelligent robot inspection method, where the method includes:
s10: generating a robot inspection task and calling digital information of an inspection area;
s20: the method comprises the steps of interactively obtaining basic data of the robot, wherein the basic data comprise size data, function data and electric quantity data;
s30: executing model initialization of a path planning model through the basic data, inputting the digital information and the inspection task into the path planning model, outputting a path planning result, and taking the path planning result as a standard reference path of the robot;
step S30 further includes the steps of:
s31: acquiring battery electric quantity data of the robot, and generating an actual electric quantity evaluation value according to an acquisition result;
s32: extracting historical environment execution data of the robot, and constructing a battery environment influence factor of the robot according to the historical environment execution data;
s33: analyzing the inspection task, and collecting environment prediction data of a task execution node;
S34: generating a battery influence coefficient according to the environment prediction data and the battery environment influence factor;
s35: initializing the electric quantity data based on the actual electric quantity evaluation value, and generating a path planning constraint of the path planning model through the battery influence coefficient.
Specifically, before the robot is put into the inspection area, according to the environment of the inspection area, an inspection task which needs to be executed by the robot is generated on the intelligent inspection system of the robot in multiple scenes, for example, the environment of the inspection area is a tunnel, the corresponding inspection task can be a tunnel inspection starting point and a tunnel inspection end point of one or more sections of tunnels, and meanwhile, digital information of the inspection area is acquired, and the digital information includes but is not limited to map data and sensor data.
The basic data comprise size data (including but not limited to height, width and length), functional data (including but not limited to bearing capacity and sensor types) and electric quantity data, the basic data of the robot are acquired through the interactive connection between the multi-scene intelligent robot inspection system and the robot, namely through the transmission and interaction of signals, a communication network is formed between the multi-scene intelligent robot inspection system and the robot, and technical support is provided for acquiring the basic data of the robot in real time.
Constructing a path planning model; the basic data is provided with a model initialization mark, a path planning model is initialized through the basic data, basic data including but not limited to the size, the function and the electric quantity of a robot and digital information including but not limited to a map, the size and a patrol point of a patrol area are input; the input of the path planning model is the digital information and the routing inspection task, the output is the path planning result, the path planning is carried out through the path planning model, namely, the robot supports routing inspection according to a certain path or a plurality of paths, the path planning result is used as a standard reference path of the robot, so that the robot is ensured to carry out routing inspection according to an expected path, and meanwhile, the automatic scene routing inspection can be realized through the application of the digital information and the path planning model, and the automation degree and the intelligent level of routing inspection are improved.
Constructing a path planning model, wherein the path planning model comprises a scene simulation modeling layer and a path automatic planning layer; specifically, a scene simulation modeling layer is arranged: using 3DStudioMax (software name), maya (software name), rhino (software name), blender (software name) or any other modeling software as a bottom modeling logic, and performing scene modeling based on the digitized information of the inspection area; setting a path automatic planning layer: and (3) comparing the routing inspection starting point, the routing inspection end point and the routing inspection passing point of the routing inspection task, and performing path planning by using a Dijkstra algorithm, wherein the continuity of the path needs to be considered in the path generation process.
And configuring residual electricity constraint of the robot, further, updating equipment characteristics according to equipment states and environment states, and redetermining residual electricity, and further, executing battery electric quantity data acquisition on the robot to obtain an acquisition result, and generating an actual electric quantity storage evaluation value according to the acquisition result, wherein the actual electric quantity storage evaluation value is equal to actual consumed electric quantity/(battery nominal capacity-battery residual electric quantity) ×100%, and directly substituting the acquisition result into the acquisition result to calculate to obtain the actual electric quantity storage evaluation value.
The historical environment execution data is related data of the environment where the robot is located in an execution task in the past year, and the historical environment execution data of the robot is extracted by searching the related data of the environment in a data storage module of the robot; the battery environment influence factor refers to the influence degree of various factors related to the battery use environment on the battery life and the electricity consumption in the historical execution data of the robot, for example, the factors such as temperature, humidity, illumination and the like may influence the performance and the life of the battery, and the battery environment influence factor may reflect the influence degree of the factors on the battery, so that a basis is provided for the battery management of the robot, and the battery environment influence factor of the robot for measuring the performance of the battery is constructed according to the historical environment execution data: generally, the historical environment execution data and the historical execution data of the robot include the battery consumption condition, battery life and other data of the robot in different environments, the historical environment execution data and the historical execution data of the robot can be subjected to data association mapping, the influence degree of the historical environment execution data and the historical execution data of the robot on the battery is obtained, an index with the influence degree not lower than 0.2% is used as a battery environment influence factor, and the battery management strategy of the robot is guided according to the battery environment influence factor, for example, battery maintenance measures are timely set in different environments, or the working plan of the robot is adjusted to fully utilize the life and energy of the battery.
Analyzing the patrol task, analyzing the patrol task into different task execution nodes, and simultaneously, obtaining the environment prediction data as the prediction environment at the moment corresponding to the task execution nodes; taking the moment corresponding to the execution node and the address information corresponding to the execution node as constraint conditions, and collecting environment prediction data of the task execution node on an online weather prediction platform; and carrying out normalization processing on the environment prediction data and the battery environment influence factors, taking the normalization processing result of the battery environment influence factors as a weight value, and carrying out weighted calculation on the normalization processing result of the environment prediction data to obtain battery influence coefficients, wherein the battery influence coefficients are used for describing the degree of environmental influence of battery performance.
In the continuous use process of the robot, taking the product of the actual electric storage capacity evaluation value multiplied by the nominal capacity of the battery as the actual electric storage capacity of the battery, initializing the electric quantity data, and generating a path planning constraint of the path planning model through the battery influence coefficient at the same time: calculating the estimated service time of the battery according to the battery influence coefficient, and converting the estimated service time into a corresponding energy consumption value; carrying out path planning according to Dijkstra algorithm, generating a path planning scheme, and calculating the energy consumption value of each path segment; comparing the energy consumption value of the battery with the energy consumption value of the path section, and judging whether the path planning task can be completed or not; if the task cannot be completed, the path planning scheme is adjusted, and the energy consumption value of the path segment is recalculated until a feasible scheme is found. The generated path planning constraints are applied to a path planning model to ensure that the path planning scheme meets the energy consumption limits of the battery.
Through collection and analysis of battery data and environment data, the energy storage condition of the battery of the robot can be evaluated, and corresponding battery influence coefficients are generated according to environment prediction data, so that more accurate constraint is provided for path planning, effective execution of the path planning is guaranteed, meanwhile, the battery utilization rate and the working efficiency of the robot can be improved, and the downtime is reduced.
As shown in fig. 2, step S35 further includes the steps of:
s351: setting a reserved electric quantity threshold value of the battery;
s352: after the digitized information and the inspection task are input into the path planning model, executing electric quantity reservation setting of the path planning model through the reserved electric quantity threshold;
s353: and executing path planning based on the path planning model after the reserved electric quantity is set.
Specifically, a reserved electric quantity threshold of a battery is set, wherein the reserved electric quantity threshold of the battery is the difference between the electric quantity required by the robot to execute the inspection task and the residual electric quantity of the battery, so that the robot is ensured to return to a charging station with enough electric quantity when the robot finishes executing the task; after the digitized information and the inspection task are used as input data and are input into a path planning model, before path planning is carried out, the electric quantity reservation setting in the path planning model is adjusted according to a set electric quantity reservation threshold value, so that the robot is ensured not to stop midway or return to a charging station due to insufficient electric quantity when executing the task; and executing path planning by using the path planning model according to the set reserved electric quantity threshold value, the input digital information and the inspection task, wherein the planned path also comprises a path for the robot to return to the charging station from the current position.
By setting the reserved electric quantity threshold value of the battery and the reserved electric quantity of the path planning model, the robot can be ensured to reserve enough electric quantity when the robot executes the inspection task, the task interruption or the battery exhaustion caused by the insufficient electric quantity is avoided, meanwhile, the working reliability and the long-time working capacity of the robot can be improved, and the successful completion of the task is ensured.
S40: controlling the robot to execute scene inspection based on the standard reference path, and executing path image acquisition of the standard reference path through an image acquisition device;
s50: performing obstacle and road surface recognition on the path image acquisition result to generate obstacle characteristics and road surface characteristics;
s60: configuring a pre-aiming distance and an obstacle avoidance reference according to the obstacle characteristics, the road surface characteristics and the basic data;
step S60 further includes the steps of:
s61: constructing a track association coefficient set of the robot and the road surface features through big data;
s62: mapping and matching the track association coefficient set based on the road surface characteristics to obtain a matching track association coefficient;
s63: and when the pretightening distance configuration is carried out, carrying out initial pretightening distance compensation through the matching track association coefficient so as to complete pretightening distance configuration.
Specifically, the path planning result is used as a reference path of the robot, and scene inspection is realized by controlling the motion trail of the robot. Meanwhile, the robot is equipped with an image acquisition device, and path image acquisition of a standard reference path can be performed through the mounted image acquisition device, namely, the robot can shoot path images in the process of traveling along the standard path so as to facilitate subsequent data processing.
The obstacle features are used for representing obstacles on a robot inspection path, such as roadblocks, vehicles and pedestrians; the road surface features are used for representing the road surface features on the robot inspection path, such as lane lines, road signs and speed reduction zones; computer vision analysis is carried out on the path image acquisition result, and obstacle characteristics and road surface characteristics on the path are identified; according to the obstacle characteristics, the road surface characteristics and the basic data, the pre-aiming distance of the robot and the obstacle avoidance reference are configured so that the robot can find the obstacle in time and make the avoidance action in the driving process, the obstacle avoidance reference can be the avoidance direction which is referred to when the obstacle is avoided, and generally, the direction which is free of the obstacle in front and abnormal in road surface information is optimized to avoid the obstacle, water is accumulated in the left front, and the avoidance direction corresponding to the obstacle avoidance reference is the right front.
Further, according to the obstacle characteristics, the road surface characteristics and the basic data, the pre-aiming distance of the robot is configured, which includes that the road surface characteristics are used for representing road surface characteristics on a robot inspection path, such as lane lines, road signs and speed reduction zones, based on big data, the robot model is used as a data screening condition, inspection paths of the same type of robots under different road surface characteristics are obtained, data arrangement is carried out according to the corresponding relation between the road surface characteristics and the inspection paths of the same type of robots, so that a track association coefficient set for constructing the robot and the road surface characteristics is obtained, for example, the road surface characteristics are the speed reduction zones, the inspection paths corresponding to 52.47% of robots are the road surface characteristics which keep the original direction and are properly decelerated, the inspection paths corresponding to 38.17% of robots are the road surface characteristics which are the road surface with the right side which are the road wheels crossed, the inspection paths corresponding to 9.36% of the robots are the road surface characteristics which are the left side which are the road surface with the right side which are the road wheels crossed, and under the other condition, the road surface characteristics are the road surface characteristics which are the speed limiting, the inspection paths corresponding to 87.4% of the robots are the road surfaces which keep the original direction and are proper deceleration, and the corresponding to the road surface characteristics, and the track association coefficient set is formed by adding 52.47.17%, the corresponding to the corresponding characteristics to the corresponding track association coefficient set of the road characteristics to the robot.
And matching the current road surface characteristics serving as matching conditions in the track association coefficient set to find out the track association coefficient most similar to the current road surface characteristics, for example, the current road surface characteristics are track association coefficients= (52.47% +87.4%)/2 corresponding to the speed reduction zone and the corresponding inspection path of the robot maintaining the original direction.
The pre-aiming distance refers to the pre-aiming distance which is needed to be overlapped on the original pre-aiming point when the road surface information is abnormal, such as water, ice and snow, and the road surface information is abnormal, so that the road surface information is abnormal; when the PP (pure tracking algorithm) algorithm is used for controlling, the pose and wheelbase of the controlled robot are all fixed information, and mainly depend on the setting of the pretightening distance: if the value of the pre-aiming distance is too small, the pre-aiming point can be continuously tracked in a wrapping way during robot tracking, so that the phenomenon of driving oscillation is generated, the stability of robot control is poor, the driving track becomes long, and the speed of the robot is difficult to effectively improve; if the pre-aiming distance is too large, the linear tracking of the controlled robot is slow, the over-bending steering is insufficient, and the phenomenon of 'shoveling a close road' occurs, so that the robot is difficult to accurately run along the track, and the tracking precision is reduced.
The initial pretightening distance can be 0.1m, when the pretightening distance configuration is carried out, initial pretightening distance compensation is carried out through the matching track association coefficient, an initial pretightening distance is overlapped on the original pretightening point towards the avoidance direction defined by the obstacle avoidance reference to serve as a new pretightening point, and whether obstacle crossing or road surface information abnormality is judged: if the road surface information is abnormal or passes over the obstacle, setting the pretightening distance to be 0.1m; if the obstacle or the road surface information is not crossed, superposing two initial pretightening distances as new pretightening points in the direction of avoiding the obstacle defined by the original pretightening points, judging whether the obstacle or the road surface information is crossed, and continuously repeating the steps until the obstacle or the road surface information is crossed, so as to complete pretightening distance configuration, wherein the pretightening distance is equal to one or more initial pretightening distances. The pre-aiming distance is configured, so that the robot can track the target point better in the running process, and the running accuracy and stability of the robot are ensured.
S70: generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path;
S80: and controlling the robot to finish the inspection task according to the regression path.
As shown in fig. 3, step S70 further includes the steps of:
s71: determining a rear axle center coordinate of the robot according to the basic data;
s72: taking the center coordinate of the rear axle as the center of a circle, taking the pretightening distance as the radius to carry out track intersection, determining the intersection point of the generated circle track and the standard reference path, and taking the intersection point as a pretightening point;
s73: taking the pre-aiming point as a target point, and executing the running track fitting of the robot;
s74: and generating the regression path according to the driving track fitting result.
Specifically, generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path: the obstacle avoidance reference can be an avoidance direction which is referred when an obstacle is avoided, a pre-aiming distance is overlapped on an original pre-aiming point towards the avoidance direction which is limited by the obstacle avoidance reference to serve as a new pre-aiming point, a PP algorithm is used for realizing initial pre-aiming distance compensation based on the geometric relation between the robot and a standard reference path, the steering radius and the running curvature are calculated in the geometric relation among the pre-aiming distance, the new pre-aiming point and the position of the robot, finally, the front wheel steering angle control quantity of the robot is calculated according to the wheelbase and the pre-aiming distance of the robot, and the robot can continuously approach a path corresponding to an obstacle or road surface information abnormal region under the action of the front wheel steering angle control quantity of the robot, so that the initial pre-aiming distance compensation is realized, and a regression path is generated; and controlling the robot to finish the inspection task according to the regression path.
Further, generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path, wherein the rear axle center coordinate refers to the coordinate of the center point of the rear axle of the robot, and determining the rear axle center coordinate of the robot according to the information disclosed in the basic data; taking the center coordinate of the rear axle as the center of a circle, taking the pre-aiming distance as the radius, and making a circle which is intersected with the standard reference path track in the running direction of the robot, determining an intersection point of the generated circle track and the standard reference path, and taking the intersection point of the circle track and the standard reference path as a pre-aiming point; obtaining an arc track of a circle drawn by taking the center coordinate of the rear axle as the circle center and the pretightening distance as the radius, taking the pretightening point as a target point, enabling the robot to reach the pretightening point according to the arc track, and executing a PP algorithm to control the running track fitting of the robot to obtain a running track fitting result; and splicing the fitting results of the multi-section driving track to obtain the regression path.
By determining the center coordinates of the rear axle and the pre-aiming distance, the intersection point of the circular track and the standard reference path can be accurately calculated, and the pre-aiming point is obtained and used as the target point of the robot. Through the running track fitting, the robot can run more accurately according to the pre-aiming point, and the accuracy of path planning and navigation is improved. The finally generated regression path can better adapt to the actual running requirement, so that the robot can effectively run on the standard reference path, and the stable and accurate navigation capability is maintained.
Step S70 further includes the steps of:
s75: generating a predetermined robot contact trajectory based on the dimensional data and the regression path;
s76: performing obstacle contact fitting through the robot contact track and the obstacle avoidance reference to generate a fitting result;
s77: judging whether obstacle avoidance contact exists in the fitting result;
s78: if obstacle avoidance contact exists, step pre-aiming distance increase fitting is carried out according to the contact range;
s79: and regenerating a running track according to the fitting result.
Specifically, the dimension data, that is, dimension information of the robot, includes length data, width data, and height data, and if the robot is a four-wheel robot, the width data includes distance data between the left tire and the right tire; if the robot is a track robot, the width data comprises distance data between tracks; the predetermined robot contact track is provided with width data in the size data, is restored according to the robot width data in the size data based on the regression path, and is measured, the predetermined robot contact track represents a track of a robot possibly contacting an obstacle in the driving process, if the avoidance direction corresponding to the avoidance reference is the right front direction, the robot is a four-wheel robot at the same time, the predetermined robot contact track is a track of a left tire contacting the obstacle, or is a track robot at the same time, the predetermined robot contact track is a track of a left side track contacting the obstacle, and coordinate points obtained by fitting the obstacle contact are combined to obtain the fitting result.
Obstacle avoidance contact refers to contact between a regression path of robot obstacle avoidance and an obstacle or abnormal road surface information, and whether the situation of contact with the obstacle exists in the fitting result is judged: if obstacle avoidance contact exists, step pre-aiming distance increase fitting is carried out according to a contact range, the corresponding pre-aiming distance is set to be 0.1m, the single increase step length corresponding to the step pre-aiming distance increase is equal to 0.1m multiplied by 10%, and if the length data corresponding to the contact range is smaller than or equal to 10% of the length data in the size information of the robot, the corresponding step pre-aiming distance is doubled single increase step length; if the length data corresponding to the contact range is smaller than or equal to 20% of the length data in the size information of the robot, the corresponding step pre-aiming distance is a double single increment step length; after finishing single increment of the pre-aiming distance, regenerating a corresponding regression path and a preset robot contact track, and judging whether obstacle avoidance contact exists according to a fitting result: if obstacle avoidance contact exists, repeating the steps until the obstacle avoidance contact does not exist; and if no obstacle avoidance contact exists, taking a regression path corresponding to the pre-aiming distance obtained by increasing the step pre-aiming distance as a regenerated running track.
And generating a preset robot contact track based on the size data and the regression path, and performing obstacle avoidance planning in advance by detecting the contact condition of the robot and an obstacle through obstacle contact fitting and judgment. If obstacle avoidance contact exists, step pre-aiming distance increase fitting is carried out according to the contact range, and single increase step length of the pre-aiming distance is adjusted so as to realize more accurate obstacle avoidance planning. Finally, the running track is regenerated according to the fitting result, so that the robot can avoid the obstacle and run safely.
The embodiment of the application further comprises the steps of:
s81: generating an accompanying power map of the standard reference path, and generating N key response nodes of the standard reference path;
s82: performing electric quantity detection of the robot on the N key response nodes to generate an electric quantity detection result;
s83: performing electric quantity verification of N key response nodes based on the electric quantity detection result and the accompanying electric quantity mapping, and generating a first electric quantity abnormal result;
s84: performing inter-node power reduction trend analysis of N key response nodes based on the power detection result and the accompanying power mapping, and generating a second power abnormal result;
s85: and performing control management on the robot according to the first electric quantity abnormal result and the second electric quantity abnormal result.
Specifically, through robot inspection and electric quantity detection, the abnormal condition of electric quantity is found fast to provide more accurate abnormal judgment and early warning through electric quantity verification and trend analysis, and then improve the efficiency and the accuracy that the robot inspected, specific step includes: before the inspection task starts, matching and mapping the electric quantity information with different positions of a path to generate an accompanying electric quantity map of the standard reference path to form a path-electric quantity distribution map, setting a key response node on the standard reference path every 100 meters, and generating N key response nodes of the standard reference path; and respectively executing the electric quantity detection of the robot on the N key response nodes, outputting an electric quantity detection result of the robot, marking the N key response nodes on the path-electric quantity distribution diagram, and simultaneously adding the detected electric quantity detection result to the path-electric quantity distribution diagram.
Performing electric quantity verification of N key response nodes based on the electric quantity detection result and the accompanying electric quantity mapping, respectively comparing and verifying whether the electric quantity is consistent or not, and if the electric quantity is inconsistent, reserving the inconsistent key response nodes into a first electric quantity abnormal result; and carrying out power down trend analysis among the nodes of the N key response nodes based on the power detection result and the accompanying power mapping, analyzing power change trend among the key response nodes corresponding to the N key response nodes, if the power change trend is abnormal and suddenly changed, reserving the key response nodes with the abnormal and suddenly changed power change trend into a second power abnormal result, combining the first power abnormal result and the second power abnormal result, carrying out control management of the robot, and carrying out power supplement, real-time path adjustment, task adjustment and state abnormality maintenance in time so as to reduce the possibility of patrol interruption accidents caused by power abnormality.
The monitoring and control of the electric quantity state of the robot can be realized by generating the accompanying electric quantity mapping and key response nodes of the standard reference path and carrying out electric quantity detection and management on the robot. Through electric quantity verification and electric quantity descending trend analysis among nodes, abnormal electric quantity conditions can be found in time, corresponding control management is carried out according to the results, the robot is ensured to normally work and keep enough electric quantity when executing tasks, and overall, the reliability and stability of the work of the robot can be improved, and the smooth completion of the tasks is ensured.
In summary, the intelligent inspection method for the robots in multiple scenes provided by the embodiment of the application has the following technical effects:
1. the method has the advantages that the robot inspection task is generated, and the digital information of the inspection area is called; the method comprises the steps of interactively obtaining basic data of the robot, wherein the basic data comprise size data, function data and electric quantity data; executing model initialization of a path planning model through basic data, inputting digital information and a patrol task into the path planning model, outputting a path planning result, and taking the path planning result as a standard reference path of the robot; controlling the robot to execute scene inspection based on the standard reference path, and executing path image acquisition of the standard reference path through the image acquisition device; performing obstacle and road surface recognition on the path image acquisition result to generate obstacle characteristics and road surface characteristics; configuring a pre-aiming distance and obstacle avoidance references according to the obstacle characteristics, the road surface characteristics and the basic data; generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path; according to the method, the requirements of actual scenes are compared, the pre-aiming point is configured, the regression path is generated, the inspection efficiency and accuracy are greatly improved, the inspection cost and risk are reduced, meanwhile, through intelligent path planning and scene recognition, the robot can efficiently execute the inspection task, the quality and reliability of inspection data are improved, and powerful supporting technical effects are provided for subsequent data analysis and decision.
2. The central coordinate of the rear axle of the robot is determined according to the basic data; taking the center coordinate of the rear axle as the center of a circle, taking the pre-aiming distance as the radius to carry out track intersection, determining the intersection point of the generated circle track and the standard reference path, and taking the intersection point as a pre-aiming point; taking the pre-aiming point as a target point, and executing the running track fitting of the robot; and generating a regression path according to the driving track fitting result. By determining the center coordinates of the rear axle and the pre-aiming distance, the intersection point of the circular track and the standard reference path can be accurately calculated, and the pre-aiming point is obtained and used as the target point of the robot. Through the running track fitting, the robot can run more accurately according to the pre-aiming point, and the accuracy of path planning and navigation is improved. The finally generated regression path can better adapt to the actual running requirement, so that the robot can effectively run on the standard reference path, and the stable and accurate navigation capability is maintained.
Example two
Based on the same inventive concept as the multi-scenario robot intelligent inspection method in the foregoing embodiment, as shown in fig. 4, an embodiment of the present application provides a multi-scenario robot intelligent inspection system, where the system includes:
The information calling module 100 is used for generating a robot inspection task and calling digital information of an inspection area;
a basic data obtaining module 200, configured to interactively obtain basic data of a robot, where the basic data includes size data, function data, and electric quantity data;
the path planning module 300 is configured to perform model initialization of a path planning model through the basic data, input the digitized information and the inspection task into the path planning model, output a path planning result, and use the path planning result as a standard reference path of the robot;
the scene inspection module 400 is configured to control the robot to perform scene inspection based on the standard reference path, and perform path image acquisition of the standard reference path through an image acquisition device;
the feature generation module 500 is configured to perform obstacle and road surface recognition on the path image acquisition result, and generate obstacle features and road surface features;
a path configuration module 600, configured to configure a pre-aiming distance and an obstacle avoidance reference according to the obstacle characteristics, the road surface characteristics, and the basic data;
a regression path generation module 700, configured to generate a regression path according to the pre-aiming distance, the obstacle avoidance reference, and the standard reference path;
And the inspection control module 800 is configured to control the robot to complete an inspection task according to the regression path.
Further, the path planning module 300 is further configured to perform the following steps:
acquiring battery electric quantity data of the robot, and generating an actual electric quantity evaluation value according to an acquisition result;
extracting historical environment execution data of the robot, and constructing a battery environment influence factor of the robot according to the historical environment execution data;
analyzing the inspection task, and collecting environment prediction data of a task execution node;
generating a battery influence coefficient according to the environment prediction data and the battery environment influence factor;
initializing the electric quantity data based on the actual electric quantity evaluation value, and generating a path planning constraint of the path planning model through the battery influence coefficient.
Further, the path planning module 300 is further configured to perform the following steps:
setting a reserved electric quantity threshold value of the battery;
after the digitized information and the inspection task are input into the path planning model, executing electric quantity reservation setting of the path planning model through the reserved electric quantity threshold;
and executing path planning based on the path planning model after the reserved electric quantity is set.
Further, the multi-scene intelligent robot inspection system is further configured to execute the following steps:
generating an accompanying power map of the standard reference path, and generating N key response nodes of the standard reference path;
performing electric quantity detection of the robot on the N key response nodes to generate an electric quantity detection result;
performing electric quantity verification of N key response nodes based on the electric quantity detection result and the accompanying electric quantity mapping, and generating a first electric quantity abnormal result;
performing inter-node power reduction trend analysis of N key response nodes based on the power detection result and the accompanying power mapping, and generating a second power abnormal result;
and performing control management on the robot according to the first electric quantity abnormal result and the second electric quantity abnormal result.
Further, the regression path generation module 700 is further configured to perform the following steps:
determining a rear axle center coordinate of the robot according to the basic data;
taking the center coordinate of the rear axle as the center of a circle, taking the pretightening distance as the radius to carry out track intersection, determining the intersection point of the generated circle track and the standard reference path, and taking the intersection point as a pretightening point;
taking the pre-aiming point as a target point, and executing the running track fitting of the robot;
And generating the regression path according to the driving track fitting result.
Further, the regression path generation module 700 is further configured to perform the following steps:
generating a predetermined robot contact trajectory based on the dimensional data and the regression path;
performing obstacle contact fitting through the robot contact track and the obstacle avoidance reference to generate a fitting result;
judging whether obstacle avoidance contact exists in the fitting result;
if obstacle avoidance contact exists, step pre-aiming distance increase fitting is carried out according to the contact range;
and regenerating a running track according to the fitting result.
Further, the path configuration module 600 is configured to perform the following steps:
constructing a track association coefficient set of the robot and the road surface features through big data;
mapping and matching the track association coefficient set based on the road surface characteristics to obtain a matching track association coefficient;
and when the pretightening distance configuration is carried out, carrying out initial pretightening distance compensation through the matching track association coefficient so as to complete pretightening distance configuration.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any of the methods to implement embodiments of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (8)
1. The intelligent inspection method for the robots in multiple scenes is characterized by comprising the following steps:
generating a robot inspection task and calling digital information of an inspection area;
the method comprises the steps of interactively obtaining basic data of the robot, wherein the basic data comprise size data, function data and electric quantity data;
executing model initialization of a path planning model through the basic data, inputting the digital information and the inspection task into the path planning model, outputting a path planning result, and taking the path planning result as a standard reference path of the robot;
controlling the robot to execute scene inspection based on the standard reference path, and executing path image acquisition of the standard reference path through an image acquisition device;
Performing obstacle and road surface recognition on the path image acquisition result to generate obstacle characteristics and road surface characteristics;
configuring a pre-aiming distance and an obstacle avoidance reference according to the obstacle characteristics, the road surface characteristics and the basic data;
generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path;
and controlling the robot to finish the inspection task according to the regression path.
2. The method of claim 1, wherein the method further comprises:
acquiring battery electric quantity data of the robot, and generating an actual electric quantity evaluation value according to an acquisition result;
extracting historical environment execution data of the robot, and constructing a battery environment influence factor of the robot according to the historical environment execution data;
analyzing the inspection task, and collecting environment prediction data of a task execution node;
generating a battery influence coefficient according to the environment prediction data and the battery environment influence factor;
initializing the electric quantity data based on the actual electric quantity evaluation value, and generating a path planning constraint of the path planning model through the battery influence coefficient.
3. The method of claim 2, wherein the method further comprises:
Setting a reserved electric quantity threshold value of the battery;
after the digitized information and the inspection task are input into the path planning model, executing electric quantity reservation setting of the path planning model through the reserved electric quantity threshold;
and executing path planning based on the path planning model after the reserved electric quantity is set.
4. The method of claim 1, wherein the method further comprises:
generating an accompanying power map of the standard reference path, and generating N key response nodes of the standard reference path;
performing electric quantity detection of the robot on the N key response nodes to generate an electric quantity detection result;
performing electric quantity verification of N key response nodes based on the electric quantity detection result and the accompanying electric quantity mapping, and generating a first electric quantity abnormal result;
performing inter-node power reduction trend analysis of N key response nodes based on the power detection result and the accompanying power mapping, and generating a second power abnormal result;
and performing control management on the robot according to the first electric quantity abnormal result and the second electric quantity abnormal result.
5. The method of claim 1, wherein the method further comprises:
Determining a rear axle center coordinate of the robot according to the basic data;
taking the center coordinate of the rear axle as the center of a circle, taking the pretightening distance as the radius to carry out track intersection, determining the intersection point of the generated circle track and the standard reference path, and taking the intersection point as a pretightening point;
taking the pre-aiming point as a target point, and executing the running track fitting of the robot;
and generating the regression path according to the driving track fitting result.
6. The method of claim 5, wherein the method further comprises:
generating a predetermined robot contact trajectory based on the dimensional data and the regression path;
performing obstacle contact fitting through the robot contact track and the obstacle avoidance reference to generate a fitting result;
judging whether obstacle avoidance contact exists in the fitting result;
if obstacle avoidance contact exists, step pre-aiming distance increase fitting is carried out according to the contact range;
and regenerating a running track according to the fitting result.
7. The method of claim 1, wherein the method further comprises:
constructing a track association coefficient set of the robot and the road surface features through big data;
mapping and matching the track association coefficient set based on the road surface characteristics to obtain a matching track association coefficient;
And when the pretightening distance configuration is carried out, carrying out initial pretightening distance compensation through the matching track association coefficient so as to complete pretightening distance configuration.
8. A multi-scenario robotic intelligent inspection system for implementing a multi-scenario robotic intelligent inspection method of any one of claims 1-7, comprising:
the information calling module is used for generating a robot inspection task and calling digital information of an inspection area;
the system comprises a basic data acquisition module, a control module and a control module, wherein the basic data acquisition module is used for interactively acquiring basic data of the robot, and the basic data comprises size data, function data and electric quantity data;
the path planning module is used for executing model initialization of a path planning model through the basic data, inputting the digital information and the inspection task into the path planning model, outputting a path planning result, and taking the path planning result as a standard reference path of the robot;
the scene inspection module is used for controlling the robot to execute scene inspection based on the standard reference path and executing path image acquisition of the standard reference path through the image acquisition device;
the feature generation module is used for executing obstacle and road surface recognition on the path image acquisition result and generating obstacle features and road surface features;
The path configuration module is used for configuring a pre-aiming distance and obstacle avoidance references according to the obstacle characteristics, the road surface characteristics and the basic data;
the regression path generation module is used for generating a regression path according to the pre-aiming distance, the obstacle avoidance reference and the standard reference path;
and the inspection control module is used for controlling the robot to finish the inspection task according to the return path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311606621.XA CN117311369B (en) | 2023-11-29 | 2023-11-29 | Multi-scene intelligent robot inspection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311606621.XA CN117311369B (en) | 2023-11-29 | 2023-11-29 | Multi-scene intelligent robot inspection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117311369A true CN117311369A (en) | 2023-12-29 |
| CN117311369B CN117311369B (en) | 2024-02-09 |
Family
ID=89274087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311606621.XA Active CN117311369B (en) | 2023-11-29 | 2023-11-29 | Multi-scene intelligent robot inspection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117311369B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117850427A (en) * | 2024-01-10 | 2024-04-09 | 苏州旷特智能科技有限公司 | Intelligent obstacle avoidance system and method based on logistics cargo handling equipment |
| CN118192618A (en) * | 2024-05-20 | 2024-06-14 | 山东瓦利斯智能科技有限公司 | Path planning method and system for electric power inspection robot |
| CN118192615A (en) * | 2024-05-16 | 2024-06-14 | 江苏三铭智达科技有限公司 | Robot path control method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109263639A (en) * | 2018-08-24 | 2019-01-25 | 武汉理工大学 | Driving path planing method based on state Grid Method |
| CN109900289A (en) * | 2019-03-18 | 2019-06-18 | 北京智行者科技有限公司 | Paths planning method and device based on closed-loop control |
| WO2021027568A1 (en) * | 2019-08-12 | 2021-02-18 | 华为技术有限公司 | Obstacle avoidance method and device |
| CN114637292A (en) * | 2022-03-14 | 2022-06-17 | 湖南大学 | Vehicle track tracking robust control method and system considering obstacle avoidance |
| CN115123310A (en) * | 2022-08-31 | 2022-09-30 | 中汽数据(天津)有限公司 | Unmanned vehicle obstacle avoidance local path planning method, device and storage medium |
-
2023
- 2023-11-29 CN CN202311606621.XA patent/CN117311369B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109263639A (en) * | 2018-08-24 | 2019-01-25 | 武汉理工大学 | Driving path planing method based on state Grid Method |
| CN109900289A (en) * | 2019-03-18 | 2019-06-18 | 北京智行者科技有限公司 | Paths planning method and device based on closed-loop control |
| WO2021027568A1 (en) * | 2019-08-12 | 2021-02-18 | 华为技术有限公司 | Obstacle avoidance method and device |
| CN114637292A (en) * | 2022-03-14 | 2022-06-17 | 湖南大学 | Vehicle track tracking robust control method and system considering obstacle avoidance |
| CN115123310A (en) * | 2022-08-31 | 2022-09-30 | 中汽数据(天津)有限公司 | Unmanned vehicle obstacle avoidance local path planning method, device and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117850427A (en) * | 2024-01-10 | 2024-04-09 | 苏州旷特智能科技有限公司 | Intelligent obstacle avoidance system and method based on logistics cargo handling equipment |
| CN118192615A (en) * | 2024-05-16 | 2024-06-14 | 江苏三铭智达科技有限公司 | Robot path control method and system |
| CN118192618A (en) * | 2024-05-20 | 2024-06-14 | 山东瓦利斯智能科技有限公司 | Path planning method and system for electric power inspection robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117311369B (en) | 2024-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117311369B (en) | Multi-scene intelligent robot inspection method | |
| CN109755995B (en) | Robot automatic charging docking method based on ROS robot operating system | |
| CN109760675B (en) | Method, device, storage medium and terminal equipment for predicting vehicle track | |
| CN107703945A (en) | A kind of intelligent farm machinery paths planning method of multiple targets fusion | |
| CN112783144B (en) | Path generation method, path planning method, system and equipment thereof | |
| CN109084798A (en) | Network issues the paths planning method at the control point with road attribute | |
| CN105404844A (en) | Road boundary detection method based on multi-line laser radar | |
| CN110986945B (en) | Local navigation method and system based on semantic altitude map | |
| US20230419824A1 (en) | Method and device for determining traffic stream information, electronic equipment and storage medium | |
| CN110806585B (en) | Robot positioning method and system based on trunk clustering tracking | |
| CN113074748B (en) | Path planning method and device for unmanned equipment | |
| CN115435798A (en) | Unmanned vehicle high-precision map road network generation system and method | |
| CN112085843B (en) | Tunnel class target feature real-time extraction and measurement method and device | |
| CN116105742B (en) | Composite scene inspection navigation method, system and related equipment | |
| CN116576857A (en) | Multi-obstacle prediction navigation obstacle avoidance method based on single-line laser radar | |
| CN113654556A (en) | Local path planning method, medium and equipment based on improved EM algorithm | |
| CN116129066A (en) | Automatic driving high-precision map model based on digital twin and high-precision static map manufacturing method | |
| CN114638103A (en) | Automatic driving joint simulation method and device, computer equipment and storage medium | |
| CN114267191A (en) | Control system, method, medium, equipment and application for relieving traffic jam of driver | |
| CN116448134B (en) | Vehicle path planning method and device based on risk field and uncertain analysis | |
| CN117168488A (en) | Vehicle path planning method, device, equipment and medium | |
| CN112699575A (en) | Method and system for measuring and calculating relative position in virtual vehicle test platform | |
| CN111951552A (en) | Method and related device for risk management in automatic driving | |
| CN114594776B (en) | Navigation obstacle avoidance method based on layering and modular learning | |
| CN116295497A (en) | Path generation method, path generation device, robot and storage medium |
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 |