CN210031639U - Unmanned sweeper - Google Patents
Unmanned sweeper Download PDFInfo
- Publication number
- CN210031639U CN210031639U CN201920154098.4U CN201920154098U CN210031639U CN 210031639 U CN210031639 U CN 210031639U CN 201920154098 U CN201920154098 U CN 201920154098U CN 210031639 U CN210031639 U CN 210031639U
- Authority
- CN
- China
- Prior art keywords
- vehicle
- ecu
- main controller
- road
- unmanned
- 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
Images
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The utility model relates to the field of sanitation vehicles, in particular to an unmanned sweeper, which comprises a sweeper body, a running system, a steering system, a road recognition system, a main controller (ECU) and a sweeping system, wherein the running system, the steering system, the road recognition system, the main controller (ECU) and the sweeping system are arranged on the sweeper body; the running system comprises a straight running motor and a straight running wheel driven by the straight running motor; the steering system comprises a steering motor and a steering wheel driven by the steering motor; the method is characterized in that: the road identification system comprises a high-definition camera for collecting road surface images in front of the vehicle and an industrial personal computer for processing image information, identifying road conditions in front of the vehicle, planning driving directions and issuing driving instructions; the industrial personal computer is controlled and connected to a main controller (ECU), and the main controller (ECU) is controlled and connected with the direct-current motor and the steering motor. The unmanned sweeper can complete the workload of a plurality of sanitation workers as long as the path for sweeping the community is set.
Description
Technical Field
The utility model relates to a sanitation vehicle field especially relates to an unmanned motor sweeper.
Background
The traditional sanitation mode is basically the following two modes: 1. the sanitation workers manually clean garbage in a community or a street; 2. the sanitation cleaning vehicle is driven manually to clean garbage in a residential area or a street. Because the sanitation system depends on manpower seriously and the former sanitation workers are not enough, the current sanitation workers need overload to finish the cleaning workload every day, and meanwhile, because the sanitation workers often work at the roadside, certain hidden danger is brought to the life safety of the sanitation workers; therefore, in order to reduce the operation and maintenance cost of sanitation, reduce the working pressure of sanitation workers and guarantee the life safety of the sanitation workers, a solution of intelligent sanitation is provided in the China sanitation industry under the current aging shadow cage of population for a long time, and the intelligent unmanned sweeper is popularized in the sanitation industry, so that the sanitation workers can enjoy the convenience and the welfare brought to the sanitation workers by unmanned driving technology and intelligent sanitation.
The utility model discloses a street sweeper is disclosed in chinese utility model patent specification CN 201635282, this kind of street sweeper includes the automobile body, the device of sweeping the floor, dust extraction, and wireless charging device, still is equipped with remote control system on the automobile body, keeps away barrier sensor, three-dimensional image recognition device, and remote control system includes camera, signal transceiver and display, installs around the automobile body outside and keeps away barrier sensor, and three-dimensional image recognition device includes camera, three-dimensional image acquisition module and image processing module, the utility model discloses can realize the unmanned control of street sweeper, also can remote control, convenient to use to possess the dust absorption function simultaneously. However, the vehicle body is provided with a remote control system, an obstacle avoidance sensor and a three-dimensional image recognition device, unmanned cleaning and path planning can be realized, but the vehicle body is more complex and has lower accuracy.
Disclosure of Invention
In order to solve the above problem, the utility model discloses a first aim at provides an unmanned sweeper, this unmanned sweeper as long as set for the route of cleaning the district, the work load of many sanitation workers just can be accomplished to unmanned sweeper, can solve the serious not enough problem of sanitation worker, simultaneously also greatly reduced sanitation worker's work load, shared sanitation worker's operating pressure.
In order to achieve the above purpose, the utility model adopts the following technical scheme:
an unmanned sweeper comprises a sweeper body, a running system, a steering system, a road recognition system, a main controller (ECU) and a sweeping system, wherein the running system, the steering system, the road recognition system, the main controller (ECU) and the sweeping system are arranged on the sweeper body; the running system comprises a straight running motor and a straight running wheel driven by the straight running motor; the steering system comprises a steering motor and a steering wheel driven by the steering motor; the method is characterized in that: the road identification system comprises a high-definition camera for collecting road surface images in front of the vehicle and an industrial personal computer for processing image information, identifying road conditions in front of the vehicle, planning driving directions and issuing driving instructions; the industrial personal computer is controlled and connected to a main controller (ECU), and the main controller (ECU) is controlled and connected with the direct-current motor and the steering motor.
Preferably, the vehicle body is further provided with an anti-collision system, the anti-collision system comprises first ultrasonic sensors arranged on two sides of the vehicle body, and the first ultrasonic sensors are used for detecting the distance between the vehicle and the road teeth in real time, converting distance signals into electric signals through A/D conversion and sending the electric signals to a main controller (ECU).
Preferably, the anti-collision system further comprises second ultrasonic sensors arranged at the head and the tail of the vehicle, and the second ultrasonic sensors are used for detecting a safe distance, converting a distance signal into an electric signal through A/D conversion, and sending the electric signal to a main controller (ECU).
Preferably, the vehicle body is further provided with an angle measuring instrument for sensing a vehicle body angle, a turning direction and a turning angle.
Preferably, the sweeping system comprises two groups of side brush assemblies respectively arranged at two sides of the frame, a rolling brush assembly arranged on the frame at the rear side of the two groups of side brush assemblies, and a dust collection assembly matched with the rolling brush assembly for use.
Preferably, the unmanned sweeping vehicle is characterized in that: the industrial computer detects the road surface image in front of the vehicle collected by the high-definition camera, and immediately sends an instruction to the main controller (ECU) after detecting garbage, and the main controller (ECU) controls and connects the side brush assembly, the rolling brush assembly and the dust collection assembly.
Preferably, a brake-by-wire system is further arranged on the vehicle body, the brake-by-wire system comprises a brake pedal module and a brake execution mechanism, a main controller (ECU) controls the brake pedal module, the brake pedal module controls the brake execution mechanism, and the brake execution mechanism acts on the straight-going wheels.
The utility model adopts the above technical scheme, this technical scheme relates to an unmanned motor sweeper. The main controller (ECU) in the unmanned sweeper receives the image information of the road identification system, processes the image information and identifies the road condition in front of the vehicle; the main controller (ECU) controls a driving system and a steering system to plan the driving direction based on the set road condition information; and the main controller (ECU) receives a distance signal between the vehicle and the road tooth fed back by the anti-collision system, and controls the driving system and the steering system to plan and adjust the distance between the vehicle and the road tooth or to adopt braking in time based on the distance signal. Therefore, the unmanned sweeper can complete the workload of a plurality of sanitation workers as long as the path for sweeping the community is set, can solve the problem that sanitation workers are seriously insufficient, greatly reduces the workload of the sanitation workers and shares the working pressure of the sanitation workers.
Drawings
Fig. 1 is a side view schematically showing the structure of an unmanned sweeping vehicle.
Fig. 2 is a front structural schematic diagram of the unmanned sweeping vehicle.
Fig. 3 is a functional block diagram of the unmanned sweeping vehicle.
Fig. 4 is a hough line transformation schematic diagram adopted in a straight road section driving scheme of the unmanned sweeper.
Fig. 5 is an execution circuit diagram of an obstacle avoiding scheme of the unmanned sweeping vehicle.
Detailed Description
The preferred embodiments of the present invention will be described in further detail with reference to the accompanying drawings.
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments.
In the description of the present invention, it is to be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore, should not be construed as limiting the present invention.
Example 1:
an unmanned sweeping vehicle as shown in fig. 1 to 3 comprises a vehicle body 1, and a driving system, a steering system, a road recognition system, an anti-collision system, a brake-by-wire system, a main controller (ECU)2 and a sweeping system which are arranged on the vehicle body 1. The running system is for performing running of the vehicle, and includes a rectilinear motion motor 31 and a rectilinear motion wheel 32, and the rectilinear motion wheel 32 is driven by the rectilinear motion motor 31. The steering system includes a steering motor 41, and a steering wheel 42 connected to an output end of the steering motor 41. The straight wheels 32 are rear wheels, and the steered wheels 42 are front wheels. The road identification system comprises a high-definition camera 51 and an industrial personal computer 52, wherein the high-definition camera 51 is used for collecting road surface images in front of the vehicle, and the industrial personal computer 52 is used for processing image information, identifying road conditions in front of the vehicle, planning driving directions and issuing driving instructions.
The anti-collision system comprises first ultrasonic sensors 61 arranged on two sides of the vehicle body 1 and second ultrasonic sensors 62 arranged on the head and tail of the vehicle; the first ultrasonic sensor 61 is used for detecting the distance between the vehicle and the road tooth in real time, the second ultrasonic sensor 62 is used for detecting the safety distance, and the distance signals are converted into electric signals by the first ultrasonic sensor 61 and the second ultrasonic sensor 62 through A/D conversion and then sent to the main controller (ECU) 2.
The industrial personal computer 52 is connected to a main controller (ECU)2, and the main controller (ECU)2 is connected to the traveling motor 31 and the steering motor 41. The main controller (ECU)2 receives a driving instruction of the road recognition system, and the main controller (ECU)2 controls the driving system and the steering system to plan the driving direction. Meanwhile, the main controller (ECU)2 receives a distance signal between the vehicle and the road tooth fed back by the anti-collision system, and the main controller (ECU)2 controls the driving system and the steering system to plan and adjust the distance between the vehicle and the road tooth based on the distance signal.
In addition, the vehicle body 1 is provided with an angle measuring instrument 7 for sensing a vehicle body angle, a turning direction, and a turning angle. During the turning process of the vehicle, the turning direction and the turning angle of the vehicle are judged through the angle measuring instrument 7, and the preset turning angle is finished.
The brake-by-wire system comprises a brake pedal module 81 and a brake actuating mechanism 82, wherein a main controller (ECU)2 controls the brake pedal module 81, the brake pedal module 81 controls the brake actuating mechanism 82, and the brake actuating mechanism 82 acts on the straight-going wheel 32; the brake foot brake pedal on the traditional vehicle is omitted in the unmanned sweeper, and the brake-by-wire system is adopted after the brake-by-wire transformation. When the road recognition system detects that an obstacle is in front of a road surface through image processing, a level signal is uploaded to send a braking instruction to the main controller (ECU)2, the main controller (ECU)2 controls the whole vehicle current to the switching value port of the driving execution module to disconnect the brake pedal module 81 after receiving the level signal instruction, and the brake pedal module 81 controls the brake execution mechanism 82 to be started, so that the unmanned sweeper is in a power-off state and stops.
The cleaning system comprises two groups of side brush assemblies 91 respectively arranged on two sides of the frame, rolling brush assemblies arranged on the frame on the rear side of the two groups of side brush assemblies, and dust collection assemblies matched with the rolling brush assemblies for use. And a main controller (ECU)2 controls and connects the side brush assembly, the rolling brush assembly and the dust collection assembly. The industrial personal computer 52 detects the road surface image in front of the vehicle collected by the high-definition camera 51, immediately sends an instruction to the main controller (ECU)2 after detecting the garbage, and the main controller (ECU)2 controls the cleaning system to work and clean.
The technical scheme relates to an unmanned sweeper, wherein a main controller (ECU)2 in the unmanned sweeper receives and processes image information of a road identification system to identify a road condition in front of a vehicle; the main controller (ECU)2 controls a driving system and a steering system to plan the driving direction based on the set road condition information; and the main controller (ECU)2 receives a distance signal between the vehicle and the road tooth fed back by the anti-collision system, and the main controller (ECU)2 controls the driving system and the steering system to plan and adjust the distance between the vehicle and the road tooth based on the distance signal. Therefore, the unmanned sweeper can complete the workload of a plurality of sanitation workers as long as the path for sweeping the community is set, can solve the problem that sanitation workers are seriously insufficient, greatly reduces the workload of the sanitation workers and shares the working pressure of the sanitation workers.
Example 2:
the present embodiment is based on the structure of the unmanned sweeping vehicle in embodiment 1, and provides a driving method of the unmanned sweeping vehicle, which includes the following steps:
step 1: the system presets a driving path which comprises a starting point position, an end point position and a driving track, the driving speed and time of a straight line section in the driving track, and the turning direction, the total turning angle and the turning times in a curve section.
Step 2: the vehicle runs according to a running path preset by the system; in the driving process, a road identification system of the vehicle acquires a road surface image in front of the vehicle through a high-definition camera, and identifies the road condition in front and plans the driving direction; the anti-collision system of the vehicle identifies the distance signal between the vehicle and the road tooth, and adjusts the distance between the vehicle and the road tooth in real time or takes braking in time.
Calculating the intersection point of two straight road teeth, recording the position of the intersection point, and planning how long the section of the route runs to the intersection point of the two road teeth according to the current speed; secondly, placing a mobile phone compass on a linear road tooth, and measuring the linear angle theta 1 of the road tooth; and placing the mobile phone compass on the other linear road tooth, measuring the linear angle theta 2 of the other road tooth, measuring the theta 1-theta 2 to obtain the turning angle of the road tooth, setting left/right turning, and setting turning for a plurality of times.
The step 2 can be divided into a straight section driving step, a general curve driving step, a special curve driving step and a driving step of detecting an obstacle, which are as follows.
The running steps of the vehicle on the straight road section are as follows:
a1, shooting a road condition video in front of a vehicle by a vehicle front high-definition camera in a road recognition system, reading a frame of video image at regular time (currently, one second is related to the vehicle speed and the response requirement), and sending the frame of video image to an industrial personal computer;
step a2, processing the read image by an industrial personal computer to obtain a right edge track of the road, and depicting a right edge by adopting a virtual edge line;
step a21, converting a picture returned by shooting into a gray-scale image by an industrial personal computer through an algorithm, improving the image quality of the picture, and enabling the display effect of the image to be clearer, finer and easy to identify;
step a22, histogram equalization is carried out on the picture, so that the gray scale range of the picture is expanded or the gray scale is uniformly distributed, thereby increasing the contrast;
step a23, performing Gaussian filtering on the image, and performing Gaussian filtering processing on the image to remove noise in the image;
step a24, cannny edge detection: identifying points with obvious brightness change in the digital image, eliminating information which is considered irrelevant in the image, and keeping important structural attributes in the image;
a25, carrying out Hough line transformation on the result of edge detection;
a26, returning a Hough line segment vector set;
a27, obtaining a straight line vector according to Hough straight line transformation;
step a28, as shown in fig. 4, obtaining an angle theta between the camera and the road tooth and a distance rh0 between the camera and the road tooth according to hough transform, calculating coordinate values of pt1 and pt2 according to the two values, and drawing a roadbed edge line according to the pt1- > pt2 coordinate and the direction.
Step a3, performing straight-line deviation correction, and keeping a straight-line driving state after the deviation correction is finished; the method comprises the following steps:
using the sum of the road edge straight line angle sum _ theta and theta equal to the sum of the road edge straight line angle sum _ theta and theta of each frame, recording the number line _ num of the road edge straight lines, recording the number frame _ num of the return photo frames, last frame angle last _ theta and wheel rotation time, and inputting the parameters;
firstly, judging whether the image is a first frame image, if so, calculating a first frame image angle flag _ theta which is sum _ theta/line _ num; each subsequent frame is compared to this frame; if the frame is not the first frame, firstly, performing straight-line rectification;
deviation rectifying: firstly, solving an average angle ave _ theta which is sum _ theta/line _ num, then transmitting the average angle to last _ theta of the previous frame, and then subtracting the average angle ave _ theta from the angle flag _ theta of the first frame image;
if the result is less than-3 degrees, the wheel needs to turn left;
if the result is greater than 3 degrees, the wheel needs to turn right;
if the result is between (-3, 3) degrees and is not greatly different from the initial state, the wheel is required to be corrected;
and keeping the straight driving state after the deviation correction is finished.
In addition, when the vehicle runs on a straight road section, a first ultrasonic sensor beside the vehicle body can detect the deviation of the road edge, and a distance signal between the first ultrasonic sensor and a road tooth is converted into an electric signal through A/D conversion and then sent to a main controller (ECU); the main controller (ECU) sets a safe threshold range, and once the safe threshold range is deviated, the front direction wheel can be rotated, so that the vehicle is recovered to be parallel to the road edge. For example, the distance detected by the first ultrasonic sensor is greater than the safe distance, and after the electric signal is fed back to a main controller (ECU), the main controller (ECU) immediately controls a steering motor to enable a front wheel to rotate to drive in a fine adjustment mode at a small angle towards the right until the distance between the front wheel and a road tooth detected by the first ultrasonic sensor is within the safe distance range, and the front wheel returns to a straight driving state in a fine adjustment mode at a small angle towards the left; when the first ultrasonic sensor detects that the distance is smaller than the safe distance range, the electric signal is fed back to a main controller (ECU), the main controller (ECU) immediately controls the steering motor to enable the front steering wheel to rotate and run at a small angle leftwards, until the first ultrasonic sensor detects that the distance between the first ultrasonic sensor and the road tooth is within the safe distance range, the main controller (ECU) controls the steering motor to rotate and run at a small angle rightwards for multiple times until the distance between the first ultrasonic sensor and the road tooth is recovered and kept straight running.
The driving steps of the vehicle on the curve section are as follows:
b1, shooting a road condition video in front of a vehicle by a vehicle front high-definition camera in the road recognition system, reading a frame of video image at regular time (currently, one second is related to the vehicle speed and the response requirement), and sending the frame of video image to an industrial personal computer;
b2, processing the read image by an industrial personal computer, judging road conditions in front of the road and issuing a driving instruction;
step b3, the industrial personal computer controls the walking motor to keep the walking motor moving forwards;
b4, when a turning road tooth is encountered, the first ultrasonic sensor detects the distance between the first ultrasonic sensor and the road tooth in real time, when the distance exceeds a set threshold range, an electric signal is transmitted to a main controller (ECU), the main controller (ECU) controls a turning motor to rotate at a small angle, after the angle measuring instrument detects that a certain angle is reached, an instruction signal is given to the main controller (ECU), the main controller (ECU) controls the turning motor to stop rotating firstly, and then controls a walking motor to walk forwards for a certain distance; after rotating for a plurality of times and moving for a certain distance in a straight line manner, the road tooth is kept parallel to the road tooth, and the detection distance of the first ultrasonic sensor is within a set distance threshold value, so that the purpose of turning and walking is achieved.
The vehicle runs on a special curve in the following steps:
b1, shooting a road condition video in front of a vehicle by a vehicle front high-definition camera in the road recognition system, reading a frame of video image at regular time (currently, one second is related to the vehicle speed and the response requirement), and sending the frame of video image to an industrial personal computer;
b2, processing the read image by an industrial personal computer, judging road conditions in front of the road and issuing a driving instruction;
step b3, a main controller (ECU) controls a walking motor based on a traveling command to enable the walking motor to keep moving forwards and straightly;
b4, after the pictures returned by the high-definition camera are judged by image processing for multiple times, no road tooth is found;
step b5, detecting a rear safe distance by a second ultrasonic sensor arranged on the rear surface of the vehicle body;
step b6, when the detected result and the road edge reach a safe range; the unmanned vehicle stops running, rotates in place in a directional manner, rotates for a certain angle each time, looks for road teeth through a picture transmitted back by the high-definition camera after rotating until the road teeth needing turning are found after rotating; if the road tooth turning device can rotate clockwise by 30 degrees, the pictures shot by the camera are transmitted back to the road recognition system, after image processing, if the road tooth to be turned is not found, the road tooth turning device continues to rotate by 30 degrees, and after 360 degrees of rotation, the road tooth to be turned is found.
B7, adjusting the direction of the vehicle head to face the road teeth, and keeping straight running after the vehicle body is adjusted; when the vehicle approaches the road kerb and the distance between the second ultrasonic sensor arranged on the vehicle head and the head of the road kerb is detected to exceed the range of the set threshold value, the vehicle head is correspondingly adjusted to keep the vehicle body and the road kerb within the range of the set safe distance;
b8, the unmanned vehicle continues to run until the second ultrasonic sensor at the rear side of the vehicle body detects that the distance between the unmanned vehicle and the road tooth is kept at a set safe distance, a main controller (ECU) gives a small-angle turning signal to a turning motor, in the turning process, an angle instrument detects that the turning angle reaches a set angle, the main controller (ECU) controls the turning motor to stop turning, and then the walking motor is controlled to walk forwards for a set distance; after the vehicle turns and walks for many times, and turns and walks again until the ultrasonic sensor detects that the vehicle body is parallel to the road teeth, the main controller (ECU) controls the walking motor to walk straight again, and therefore the turning of a special curve is completed.
The driving steps of the vehicle when the obstacle is detected are as follows:
step c1, the high-definition camera shoots the road condition video in front of the vehicle, and reads a frame of video image at regular time (currently, one second, which is related to the vehicle speed and the response requirement) and sends the video image to the industrial personal computer; after image processing, judging that the road surface has obstacles and the distance of the object detected by the second ultrasonic sensor arranged in front of the vehicle head is within the set alarm threshold distance;
step c2, the industrial personal computer issues an instruction, the main controller (ECU) controls the brake-by-wire system to adopt braking, and the unmanned vehicle stops;
step c3, the high-definition camera still takes pictures in the next period of time and transmits the pictures to the industrial personal computer, and the industrial personal computer judges whether the obstacles are removed; if the obstacle is judged to be cleared, the industrial personal computer issues a walking instruction, and after receiving the instruction, the main controller (ECU) controls the walking motor to start and continue walking; if the obstacles are detected for many times and not cleared, the industrial personal computer issues a traveling motor reverse rotation instruction, the unmanned vehicle backs up, when the reversing reaches a set safe distance, a main controller (ECU) controls a turning motor to rotate for a certain angle, and if the distance between the first ultrasonic sensor around the vehicle body and the obstacles is detected to be within a safe range, the vehicle continues to rotate until the vehicle can bypass the obstacles and continues to travel forwards; if the first ultrasonic sensor around the vehicle body detects that the distance between the first ultrasonic sensor and the obstacle is within the set alarm threshold range, the vehicle continues to back to the safe distance, and then the vehicle can continue to drive forwards by bypassing the obstacle. As shown in fig. 5, an infrared photoelectric sensing switch is connected to the front end of the vehicle body, and is connected to an electric quantity acquisition end of a main controller (ECU), when a photoelectric switch is triggered, a signal output end outputs a low-level signal, when the photoelectric switch is not triggered, the signal output end is a high-level signal, when the photoelectric switch is a low-level signal, a photodiode on a control board is switched on, and the low-level signal is output to a control board chip, and after the control board chip judges the low-level signal, a switching value port of a vehicle driving module is controlled to be disconnected, so that the power failure of the unmanned vehicle is stopped for 10S, then the switching value port of the reversing vehicle is controlled to be closed, so that the unmanned vehicle backs for 4S, then a turning motor is controlled to turn left for 20S, and then the unmanned vehicle runs forward, at this time, no obstacle is blocked in front of. Until the obstacle is bypassed, the road tooth is found and the vehicle continues to drive forward according to the road tooth.
Operation of the cleaning system
The unmanned sweeper with the structure can also be used for inspecting road surface garbage and controlling the cleaning system to work in real time. The method comprises the following steps: firstly, photographing through a camera to collect and construct a sample set, namely classifying the garbage on the road surface into a positive sample through collection and training; and classifying the road surface without garbage into a negative sample through collection and training. Then, the high definition camera gathers the road surface image in the front of the vehicle, and the industrial computer compares the road surface image in the front with the sample set. When the garbage is detected, an instruction is immediately issued to a main controller (ECU), and the main controller (ECU) controls the cleaning system to work and clean.
Specifically, a sample set is constructed by obtaining street pictures and marking garbage areas, and the CNN convolutional neural network is trained. The method comprises the steps of registering a real-time street picture to be detected with a clean street picture and carrying out pixel level subtraction, acquiring a recommended region by using a selective search on the subtracted picture by adopting an RCNN algorithm to obtain positions where garbage is likely to occur, inputting image data in each region where garbage is likely to occur on the real-time picture into a trained CNN convolutional neural network for recognition, judging whether the recommended frame is garbage or not according to an output vector of the CNN convolutional neural network, and marking the real-time picture if the recommended frame is garbage.
The marking means recording the upper left corner and the lower right corner of the local image with the garbage in the image. Constructing the sample set means that the marked garbage area in the image is cut out to be used as a positive sample, meanwhile, a typical non-garbage area in the image is cut out to be used as a negative sample, and the positive and negative sample pictures are normalized to be the same size.
Wherein the selective search target area parameter is set between 100 and 1000, the target aspect ratio threshold is set to 3, and the gaussian filter parameter is set to 1.0.
In specific operation, 5000 street pictures (including streets with garbage and streets without garbage) are taken under different weather and street scenes, wherein the types of garbage comprise conventional household garbage such as fruit peels, plastic bottles, plastic bags, paper scraps and the like, and the occurrence frequency of each type of garbage is more than 100 times. And after the picture is shot and marked, intercepting positive and negative samples from the picture, wherein 2000 positive samples and 4000 negative samples are obtained, and normalizing the length and the width of the intercepted image to 61 to complete the construction of a sample library.
The unmanned sweeper has the effects that:
1. high cleaning efficiency
The unmanned sweeper can complete the workload of a plurality of sanitation workers as long as the unmanned sweeper is fully charged in advance and sets a path for sweeping a cell, so that the problem of serious shortage of sanitation workers can be solved, the workload of the sanitation workers is greatly reduced, and the working pressure of the sanitation workers is shared;
2. low maintenance cost
The parts of the unmanned sweeper are damaged, the parts are replaced, and the maintenance cost is low;
3. the life safety of sanitation workers is ensured
The unmanned sweeper replaces the roadside operation of the sanitation worker, so that the life safety of the sanitation worker is effectively guaranteed;
4. diversification is brought to the structure of the personnel in the current aging sanitation industry;
the above, only be the concrete implementation of the preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto, and any person skilled in the art is in the technical scope of the present invention, according to the technical solution of the present invention and the utility model, the concept of which is equivalent to replace or change, should be covered within the protection scope of the present invention.
Claims (7)
1. The unmanned sweeper comprises a sweeper body, and a running system, a steering system, a road identification system, a main controller (ECU) and a sweeping system which are arranged on the sweeper body; the running system comprises a straight running motor and a straight running wheel driven by the straight running motor; the steering system comprises a steering motor and a steering wheel driven by the steering motor; the method is characterized in that: the road identification system comprises a high-definition camera for collecting road surface images in front of the vehicle and an industrial personal computer for processing image information, identifying road conditions in front of the vehicle, planning driving directions and issuing driving instructions; the industrial personal computer is controlled and connected to a main controller (ECU), and the main controller (ECU) is controlled and connected with the direct-current motor and the steering motor.
2. The unmanned sweeping vehicle of claim 1, wherein: the anti-collision system is further arranged on the vehicle body and comprises first ultrasonic sensors arranged on two sides of the vehicle body, the first ultrasonic sensors are used for detecting the distance between the vehicle and the road teeth in real time, and converting distance signals into electric signals through A/D conversion and sending the electric signals to a main controller (ECU).
3. The unmanned sweeping vehicle of claim 2, wherein: the anti-collision system further comprises second ultrasonic sensors arranged at the head and the tail of the vehicle, and the second ultrasonic sensors are used for detecting the safe distance and converting distance signals into electric signals to a main controller (ECU) through A/D conversion.
4. The unmanned sweeping vehicle of claim 1, wherein: the car body is also provided with an angle measuring instrument for sensing the angle, the turning direction and the turning angle of the car body.
5. The unmanned sweeping vehicle of claim 1, wherein: the sweeping system comprises two groups of side brush assemblies respectively arranged on two sides of the frame, a rolling brush assembly arranged on the frame behind the two groups of side brush assemblies, and a dust collection assembly matched with the rolling brush assembly for use.
6. The unmanned sweeping vehicle of claim 5, wherein: the industrial computer detects the road surface image in front of the vehicle collected by the high-definition camera, and immediately sends an instruction to the main controller (ECU) after detecting garbage, and the main controller (ECU) controls and connects the side brush assembly, the rolling brush assembly and the dust collection assembly.
7. The unmanned sweeping vehicle of claim 1, wherein: the bicycle is characterized in that the bicycle body is also provided with a brake-by-wire system, the brake-by-wire system comprises a brake pedal module and a brake actuating mechanism, a main controller (ECU) controls the brake pedal module, the brake pedal module controls the brake actuating mechanism, and the brake actuating mechanism acts on the straight-going wheel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920154098.4U CN210031639U (en) | 2019-01-29 | 2019-01-29 | Unmanned sweeper |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920154098.4U CN210031639U (en) | 2019-01-29 | 2019-01-29 | Unmanned sweeper |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN210031639U true CN210031639U (en) | 2020-02-07 |
Family
ID=69352671
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201920154098.4U Active CN210031639U (en) | 2019-01-29 | 2019-01-29 | Unmanned sweeper |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN210031639U (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111580515A (en) * | 2020-05-07 | 2020-08-25 | 苏州品坤智能科技有限公司 | Obstacle avoidance system for unmanned sweeping machine and obstacle avoidance method thereof |
| CN112227283A (en) * | 2020-09-24 | 2021-01-15 | 广东杜尼智能机器人工程技术研究中心有限公司 | Fall-proof unmanned sweeper |
| CN113152345A (en) * | 2021-03-26 | 2021-07-23 | 中国矿业大学 | Intelligent cleaning vehicle for fallen leaves on road surfaces of garden and non-motor vehicle lane |
| CN114485708A (en) * | 2022-01-18 | 2022-05-13 | 上汽通用五菱汽车股份有限公司 | Unmanned vehicle path planning method, system, electronic device and storage medium |
| CN114808810A (en) * | 2022-04-12 | 2022-07-29 | 吉林大学 | Semi-autonomous interactive type cleaning system and cleaning method for unmanned sweeper |
-
2019
- 2019-01-29 CN CN201920154098.4U patent/CN210031639U/en active Active
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111580515A (en) * | 2020-05-07 | 2020-08-25 | 苏州品坤智能科技有限公司 | Obstacle avoidance system for unmanned sweeping machine and obstacle avoidance method thereof |
| CN111580515B (en) * | 2020-05-07 | 2023-02-03 | 苏州品坤智能科技有限公司 | Obstacle avoidance system for unmanned sweeping machine and obstacle avoidance method thereof |
| CN112227283A (en) * | 2020-09-24 | 2021-01-15 | 广东杜尼智能机器人工程技术研究中心有限公司 | Fall-proof unmanned sweeper |
| CN113152345A (en) * | 2021-03-26 | 2021-07-23 | 中国矿业大学 | Intelligent cleaning vehicle for fallen leaves on road surfaces of garden and non-motor vehicle lane |
| CN114485708A (en) * | 2022-01-18 | 2022-05-13 | 上汽通用五菱汽车股份有限公司 | Unmanned vehicle path planning method, system, electronic device and storage medium |
| CN114808810A (en) * | 2022-04-12 | 2022-07-29 | 吉林大学 | Semi-autonomous interactive type cleaning system and cleaning method for unmanned sweeper |
| CN114808810B (en) * | 2022-04-12 | 2022-12-30 | 吉林大学 | Semi-autonomous interactive type cleaning system and cleaning method for unmanned sweeper |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109797691B (en) | Unmanned sweeper and driving method thereof | |
| CN210031639U (en) | Unmanned sweeper | |
| CN111399505B (en) | Mobile robot obstacle avoidance method based on neural network | |
| CN109460709B (en) | RTG visual barrier detection method based on RGB and D information fusion | |
| CN106680290B (en) | Multifunctional detection vehicle in narrow space | |
| CN103473951B (en) | A kind of Automatic Carpark Control System based on telepresenc | |
| CN102188311B (en) | Embedded visual navigation control system and method of intelligent wheelchair | |
| CN107745710B (en) | A kind of automatic parking method and system based on machine vision and machine learning | |
| CN203063940U (en) | Pilotless automobile device | |
| CN105759820A (en) | Road autonomous cleaning control system and method based on laser and vision | |
| CN205644275U (en) | Unmanned intelligent vehicle | |
| CN101016052A (en) | Warning method and system for preventing deviation for vehicle on high standard highway | |
| WO2020114019A1 (en) | Sanitation intelligent cleaning and sweeping operation robot control system and cleaning machinery | |
| CN112258668A (en) | Method for detecting roadside vehicle parking behavior based on high-position camera | |
| CN112477533B (en) | Dual-purpose transport robot of facility agriculture rail | |
| CN113805571B (en) | Robot walking control method, system, robot and readable storage medium | |
| CN107831760A (en) | Robot barrier thing processing system and method | |
| CN111142524A (en) | Garbage picking robot, method and device and storage medium | |
| CN111218911B (en) | Energy-saving system and energy-saving method applied to road sweeping cleaning vehicle | |
| Rossi et al. | Real-time lane detection and motion planning in raspberry pi and arduino for an autonomous vehicle prototype | |
| CN112720408B (en) | Visual navigation control method for all-terrain robot | |
| CN114495066A (en) | Method for assisting backing | |
| CN208092975U (en) | The system for realizing semiclosed parking lot management based on long short focus camera | |
| CN106960193A (en) | A kind of lane detection apparatus and method | |
| CN211933898U (en) | Cleaning robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |