Target detection system of automatic driving sanitation vehicle

Abstract

The invention provides an automatic driving sanitation vehicle target detection system, which relates to the technical field of image detection and comprises a vehicle frame for bearing a hardware layer and a software layer. The software layer comprises a laser radar, a millimeter wave radar, an ultrasonic radar and a data processing unit, the hardware layer comprises a detection mechanism and a mounting frame, the detection mechanism comprises a rotating plate and an image sensor, and the image sensor collects images through an RGB or infrared camera and judges the target category in combination with an image recognition technology. According to the scheme, a multi-sensor fusion technology is adopted, the advantages of an image sensor, a laser radar, a millimeter wave radar and an ultrasonic radar are integrated, a target can be accurately detected under different environmental conditions, and the precision and reliability of target detection are improved; and meanwhile, the acquired image data can be quickly and accurately processed through a data processing unit by adopting a fusion algorithm and a target recognition model based on deep learning, so that the recognition and classification of various targets are realized.

Description

Technical Field

[0001] This invention relates to the field of image detection, and more particularly to a target detection system for autonomous sanitation vehicles. Background Technology

[0002] With the acceleration of urbanization, the demand for urban environmental sanitation management is increasing. Traditional sanitation operations mainly rely on manual operation, which has problems such as low efficiency, high labor intensity and poor safety. The development of autonomous driving technology has provided a solution to these problems, and autonomous sanitation vehicles have emerged.

[0003] Target detection is one of the key technologies for autonomous navigation and operation of sanitation vehicles. Its performance directly affects the safety and efficiency of sanitation vehicle operations. Currently, target detection devices for autonomous sanitation vehicles mostly use a single sensor, such as a camera or lidar.

[0004] In existing technologies, single sensors have limitations in complex environments. Cameras are easily affected by factors such as lighting and weather, and their detection accuracy decreases in environments such as strong light, heavy rain, and heavy fog. Although lidar is not affected by lighting, it is insufficient in perceiving the texture and color information of the target, and it is also costly.

[0005] Therefore, it is necessary to provide a target detection system for autonomous sanitation vehicles to solve the above-mentioned technical problems. Summary of the Invention

[0006] This invention provides a target detection system for autonomous sanitation vehicles, which solves the problem that single sensors have limitations in complex environments, and that cameras are easily affected by factors such as lighting and weather, resulting in decreased detection accuracy in environments such as strong light, heavy rain, and dense fog.

[0007] To solve the above-mentioned technical problems, the present invention provides an autonomous driving sanitation vehicle target detection system, comprising a vehicle frame supporting the hardware layer and a software layer; The software layer includes lidar, millimeter-wave radar, ultrasonic radar, and a data processing unit. The hardware layer includes a detection mechanism and a mounting frame. The detection mechanism includes a rotating plate and an image sensor. The image sensor acquires images through an RGB or infrared camera and combines image recognition technology to determine the target category, distinguishing between "recyclable waste," "construction waste," "pedestrians," and "traffic signs." LiDAR acquires the three-dimensional coordinates, contours, and distance information of targets by emitting laser point clouds, identifies the position and movement trajectory of garbage, low obstacles, pedestrians, and vehicles, and forms a three-dimensional image; The image sensor, lidar, millimeter-wave radar, and ultrasonic radar transmit the acquired image data to the data processing unit. The data processing unit integrates, processes, and transmits the various image data and performs basic target detection and recognition, including visual algorithms and lidar algorithms. Visual algorithms: Deep learning algorithms such as YOLO and Faster R-CNN are used to detect images. By training a "sanitation scene-specific dataset", the recognition rate of special targets is improved. LiDAR algorithm: Use algorithms such as PointNet and PointPillars to classify the clustered point cloud, identify targets such as "pedestrians, vehicles, garbage, and obstacles", and output three-dimensional image data such as the position, size and movement speed of the targets.

[0008] Preferably, a mounting bracket is bolted to the outside of the vehicle frame. Two sliding rods are installed inside the mounting bracket. An upper sliding frame and a lower sliding frame are slidably connected to the outside of the two sliding rods. A mounting plate is fixed between the upper sliding frame and the lower sliding frame. The detection mechanism also includes a first positioning plate and a second positioning plate fixed to the outer wall of the mounting plate. A drive motor is mounted outside the mounting plate and above the first positioning plate. A lead screw is connected to the output shaft of the drive motor via a keyway. Limiting rods are provided outside the mounting plate and on both sides of the lead screw. A fixed rack is bolted to the outer wall of the mounting plate and on one side of the lead screw. A movable plate is threaded to the outer wall of the lead screw. A rotating gear is rotatably connected to the outer wall of the movable plate. The rotating plate is bolted to the outer wall of the rotating gear. The image sensor is embedded inside the rotating plate. A linkage plate is bolted to the side wall of the movable plate.

[0009] Preferably, the upper and lower ends of the limiting rod are fixedly connected to the first positioning plate and the second positioning plate, and the two ends of the lead screw are rotatably connected to the first positioning plate and the second positioning plate.

[0010] Preferably, the movable plate and the limiting rod are slidably connected and can move up and down along the vertical direction of the limiting rod, and the rotating gear and the fixed rack mesh with each other.

[0011] Preferably, it also includes auxiliary mechanisms; The auxiliary mechanism includes a first limiting plate and a second limiting plate. The first limiting plate and the second limiting plate are respectively fixed to the outer wall of the mounting plate and located below the rotating plate. A rotating ring is rotatably connected inside the second limiting plate. A toothed ring is coaxially fixed to the side wall of the rotating ring. A cleaning roller is rotatably connected inside the first limiting plate. A ratchet is connected to the axis of the cleaning roller and located inside the rotating ring via a keyway. A positioning seat is fixed to the inner wall of the rotating ring. A ratchet is installed inside the positioning seat via a torsion spring.

[0012] Preferably, a movable rack is fixedly provided at the bottom of the linkage plate, and the movable rack and the toothed ring mesh with each other.

[0013] Preferably, the ratchet and the outer wall of the ratchet are meshed, the swivel is hollow at its center, and the swivel does not contact the ratchet.

[0014] Preferably, side plates are bolted to both sides of the mounting plate, a rotating frame is rotatably mounted on the outer wall of the mounting frame via a torsion spring, a guide wheel is rotatably connected to the inner wall of the rotating frame, and a side camera is mounted on the outer wall of the rotating frame in the same horizontal direction as the guide wheel.

[0015] Preferably, the width of the guide wheel is equal to the width of the side plate, and the side of the side plate has an inclined surface structure.

[0016] Compared with related technologies, the target detection system for autonomous sanitation vehicles provided by this invention has the following advantages: Employing multi-sensor fusion technology, which integrates the advantages of image sensors, lidar, millimeter-wave radar, and ultrasonic radar, this technology can accurately detect targets under different environmental conditions, improving the accuracy and reliability of target detection. Simultaneously, the acquired image data can be processed by the data processing unit using deep learning-based fusion algorithms and target recognition models to quickly and accurately process multi-sensor data, enabling the identification and classification of various targets and the calculation of relevant target information. This provides a reliable decision-making basis for the autonomous navigation and operation of autonomous sanitation vehicles. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0018] Figure 1 The optimal structural schematic diagram provided for this invention; Figure 2 This is a schematic diagram of the software layer target image detection process provided by the present invention; Figure 3 for Figure 1 The enlarged structural diagram at point A is shown below; Figure 4 for Figure 1 The enlarged structural diagram at point B is shown below; Figure 5 for Figure 1 The enlarged structural diagram at point C is shown below; Figure 6 A schematic diagram of the initial working state of the detection mechanism and auxiliary mechanism provided by the present invention; Figure 7 for Figure 6 The diagram shows the flip-over working state of the testing mechanism. Figure 8 for Figure 6 The diagram shows a detailed structural schematic of the auxiliary mechanism. Figure 9 for Figure 8 The diagram shows a side view of the structure. Figure 10 A schematic diagram of the working state of the auxiliary mechanism provided by the present invention; Figure 11 for Figure 1 The diagram shows the working state of the rotating frame flipping when the mounting plate moves left and right.

[0019] Explanation of icon numbers: 1. Frame; 2. Mounting bracket; 3. Slide bar; 4. Mounting plate; 5. Detection mechanism; 51. Fixed rack; 52. Drive motor; 53. Lead screw; 54. Limiting rod; 55. First positioning plate; 56. Moving plate; 57. Rotary gear; 58. Rotating plate; 59. Linkage plate. 510. Image sensor; 511. Second positioning plate; 6. Auxiliary mechanism; 61. First limiting plate; 62. Second limiting plate; 63. Moving rack; 64. Rotary ring; 65. Gear ring; 66. Positioning seat; 67. Ratchet; 68. Ratchet tooth; 69. Cleaning roller; 7. Install the upper carriage; 8. Side plate, 9. Rotating frame, 10. Guide wheel, 11. Side camera, 12. Slide rack. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention provides a target detection system for autonomous sanitation vehicles.

[0022] First embodiment: Please see Figures 1 to 2 An autonomous driving sanitation vehicle target detection system includes a vehicle frame 1 that carries the hardware layer and a software layer; The software layer includes a lidar, a millimeter-wave radar, an ultrasonic radar, and a data processing unit. The hardware layer includes a detection mechanism 5 and a mounting frame 2. The detection mechanism 5 includes a rotating plate 58 and an image sensor 510. The image sensor 510 acquires images through an RGB or infrared camera and uses image recognition technology to determine the target category, distinguishing between "recyclable waste," "construction waste," "pedestrians," and "traffic signs." LiDAR acquires the three-dimensional coordinates, contours, and distance information of targets by emitting laser point clouds, identifies the position and movement trajectory of garbage, low obstacles, pedestrians, and vehicles, and forms a three-dimensional image; The image sensor 510, lidar, millimeter-wave radar, and ultrasonic radar transmit the acquired image data to the data processing unit. The data processing unit integrates, processes, and transmits the various image data, and performs basic target detection and recognition, including visual algorithms and lidar algorithms. Visual algorithms: Deep learning algorithms such as YOLO and Faster R-CNN are used to detect images. By training a "sanitation scene-specific dataset", the recognition rate of special targets is improved. LiDAR algorithm: Use algorithms such as PointNet and PointPillars to classify the clustered point cloud, identify targets such as "pedestrians, vehicles, garbage, and obstacles", and output three-dimensional image data such as the position, size and movement speed of the targets.

[0023] Please see Figure 2 The final image data of the target detection can be used to build network model data, which can then be transmitted to the cloud for target image processing, and can also help to adjust the detection in the later stage.

[0024] This embodiment: Employing multi-sensor fusion technology, which integrates the advantages of image sensor 510, lidar, millimeter-wave radar, and ultrasonic radar, this technology can accurately detect targets under different environmental conditions, improving the accuracy and reliability of target detection. At the same time, the acquired image data can be processed by the data processing unit using a deep learning-based fusion algorithm and target recognition model to quickly and accurately process multi-sensor data, enabling the identification and classification of various targets and the calculation of relevant target information. This provides a reliable decision-making basis for the autonomous navigation and operation of autonomous sanitation vehicles.

[0025] Second embodiment: Please see Figure 1 , Figures 3 to 7The frame 1 is bolted to the outside of a mounting bracket 2. Two sliding rods 3 are installed inside the mounting bracket 2. An upper sliding frame 7 and a lower sliding frame 12 are slidably connected to the outside of each sliding rod 3. A mounting plate 4 is fixed between the upper sliding frame 7 and the lower sliding frame 12. The detection mechanism 5 also includes a first positioning plate 55 and a second positioning plate 511 fixed to the outer wall of the mounting plate 4. A drive motor 52 is mounted outside the mounting plate 4 and above the first positioning plate 55. A lead screw 53 is keyway-connected to the output shaft of the drive motor 52. Limiting rods 54 are provided on both sides of the lead screw 53 outside the mounting plate 4. A fixed rack 51 is bolted to the outer wall of the mounting plate 4 and to one side of the lead screw 53. A moving plate 56 is threaded to the outer wall of the lead screw 53. A rotating gear 57 is rotatably connected to the outer wall of the moving plate 56. A rotating plate 58 is bolted to the outer wall of the rotating gear 57. An image sensor 510 is embedded inside the rotating plate 58. A linkage plate 59 is bolted to the side wall of the moving plate 56.

[0026] Please see Figure 6 In the initial state, the image sensor 510 is located in the exact center of the mounting plate 4, so normal image acquisition and detection can be performed; Please see Figure 7 When the user starts the drive motor 52, it drives the lead screw 53 to rotate. When the lead screw 53 rotates, it can control the vertical lifting and lowering of the moving plate 56. During the lifting and lowering process of the moving plate 56, the movement direction of the moving plate 56 can be restricted by the limit rod 54, thereby making the lifting and lowering movement of the moving plate 56 more stable. When the moving plate 56 drives the rotating gear 57 to move to the position of the fixed rack 51, the rotating gear 57 will mesh with the fixed rack 51 and drive the rotating plate 58 to make the image sensor 510 inside rotate. Thus, the lifting and rotation linkage can be guaranteed during the lifting and lowering process of the moving plate 56.

[0027] The upper and lower ends of the limiting rod 54 are fixedly connected to the first positioning plate 55 and the second positioning plate 511, and the two ends of the lead screw 53 are rotatably connected to the first positioning plate 55 and the second positioning plate 511.

[0028] The movable plate 56 and the limiting rod 54 are slidably connected and can be raised and lowered along the vertical direction of the limiting rod 54. The rotating gear 57 and the fixed rack 51 mesh with each other.

[0029] Understandable: Please refer to Figure 1 Mounting bracket 2 is installed on the vehicle frame 1 and is used only for target detection. Users can adjust or remove mounting bracket 2 according to actual needs. Please see Figure 1 and Figure 6Furthermore, the user can slide the upper slide 7 and the lower slide 12 along the horizontal direction of the slide bar 3, thereby changing the horizontal direction of the mounting plate 4, thus improving the flexibility of target detection.

[0030] This embodiment: The image sensor 510 adopts a lifting and rotating design, which can expand the field of view coverage, reduce detection blind spots, cover high and low dimension blind spots, detect low obstacles, detect distant targets in advance, and reserve more decision-making reaction time for autonomous sanitation vehicles. When rotating, it can achieve detection without blind spots. The accuracy of target detection depends on high-quality image input. The lifting and rotating design can optimize the imaging quality by adjusting the position and angle of the camera. Target detection in autonomous sanitation vehicles relies on multi-sensor fusion. The image sensor 510 in this case can serve as a "supplementary redundancy to the visual sensor" and complement other sensors. It can also dynamically avoid "interference from harsh environments" and ensure the stability of the detection system. Sanitation operations often face harsh environments such as rain, snow, dust, and strong light reflection. Fixed cameras are easily interfered with due to "fixed viewing angle". The image sensor 510 in this case can dynamically avoid interference sources.

[0031] Third embodiment: Please see Figures 6 to 10 It also includes auxiliary mechanisms 6; The auxiliary mechanism 6 includes a first limiting plate 61 and a second limiting plate 62. The first limiting plate 61 and the second limiting plate 62 are respectively fixed to the outer wall of the mounting plate 4 and located below the rotating plate 58. A rotating ring 64 is rotatably connected inside the second limiting plate 62. A toothed ring 65 is coaxially fixed to the side wall of the rotating ring 64. A cleaning roller 69 is rotatably connected inside the first limiting plate 61. A ratchet 67 is connected to the axis of the cleaning roller 69 and located in the keyway inside the rotating ring 64. A positioning seat 66 is fixed to the inner wall of the rotating ring 64. A ratchet 68 is installed inside the positioning seat 66 by a torsion spring.

[0032] The bottom of the linkage plate 59 is fixedly provided with a movable rack 63, and the movable rack 63 and the toothed ring 65 mesh with each other.

[0033] Please see Figure 6 and Figure 10 When the rotating plate 58 switches to the initial state from the second embodiment, the downward movement of the rotating plate 58 can be controlled. During the downward movement of the rotating plate 58, the linkage plate 59 can be linked to drive the moving rack 63 to move downward. During the downward movement of the moving rack 63, it will engage the transmission gear ring 65 to rotate counterclockwise. Please see Figure 8 and Figure 9When the toothed ring 65 rotates counterclockwise, it can synchronously drive the rotating ring 64 to rotate counterclockwise. The rotating ring 64 controls the positioning seat 66 to rotate counterclockwise, which in turn controls the ratchet 68 to rotate, thus affecting the ratchet 67 to rotate. This allows the ratchet 67 to control the rotation of the cleaning roller 69. When the image sensor 510 descends and passes through the rotating cleaning roller 69, the rotating cleaning roller 69 can wipe the lens on the surface of the image sensor 510. Understandably, since the ratchet 67 and the rotating ring 64 are connected by the ratchet 68, when the moving rack 63 rises to reset, the toothed ring 65 rotates clockwise. During the clockwise rotation, the ratchet 68 will rotate clockwise. The clockwise rotation of the ratchet 68 will not affect the rotation of the ratchet 68. Therefore, when the image sensor 510 rises for cleaning, it automatically cancels the rotation of the cleaning roller 69.

[0034] This embodiment: By incorporating auxiliary mechanism 6, image occlusion or blurring can be eliminated, preventing missed detection of targets. The working environment of sanitation vehicles is complex, and pollutants easily adhere to the surface of the image sensor 510 lens. These substances can directly obscure or blur the image, causing the target detection system to fail to see key targets. If the lens edge is obscured by pollutants, it may obscure pedestrians and non-motorized vehicles on the side of the vehicle, increasing the risk of collision. During rainy weather, mud and water splash onto the lens surface and form a "water film." Wiping and cleaning can thoroughly remove these pollutants, allowing the lens to restore its "unobstructed, high-definition" imaging capability, ensuring that targets such as ground garbage, pedestrians, and obstacles are fully presented in the image, reducing missed detections from the source. Wiping and cleaning can remove interference sources such as water stains and oil stains, allowing the image to restore normal reflection, color and contrast, reducing noise interference to the algorithm, lowering the false detection rate, and ensuring that the system only reacts to real targets.

[0035] Fourth embodiment: Please combine Figure 1 and Figure 11 Side plates 8 are bolted to both sides of the mounting plate 4. A rotating frame 9 is rotatably mounted on the outer wall of the mounting frame 2 via a torsion spring. A guide wheel 10 is rotatably connected to the inner wall of the rotating frame 9. A side camera 11 is mounted on the outer wall of the rotating frame 9 and in the same horizontal direction as the guide wheel 10.

[0036] The width of the guide wheel 10 is equal to the width of the side plate 8, and the side of the side plate 8 has a sloping structure.

[0037] In this embodiment: During the operation of the second and third embodiments, when the user slides the mounting plate 4, whether the side plates 8 on both sides of the mounting plate 4 move to the left or the right, when the inclined surface on the side plate 8 moves to the position of the guide wheel 10, it will abut against the force control guide wheel 10, thereby realizing the entire rotating frame 9 to form a flipping motion from the horizontal direction. During the flipping process, the angle of the side camera 11 is switched, and the side camera 11 is switched from the horizontal visual angle to the flipped visual angle.

[0038] This is understandable, and the use of torsion springs for rotation installation ensures that the rotating frame 9 can be reset to the horizontal angle after the angle flip is completed.

[0039] This embodiment: By setting side cameras 11 on both sides, the low blind spots on the near side can be covered. By flipping the side cameras 11 downwards, low obstacles or garbage on the near side ground can be clearly captured, avoiding the risk of sanitation vehicles scraping or running over them from the side. It can also cover the extended blind spot in front of the side. When sanitation vehicles turn or change lanes, non-motorized vehicles or pedestrians may appear in front of the side. The flip-type side cameras 11 can capture targets in front of the side in advance, leaving enough time for the system to avoid collisions and reducing the risk of turning collisions. It can also adapt to the needs of multiple operation scenarios, improve the scenario flexibility of target detection. The detection focus of the side camera 11 is different for different scenarios, and the flip angle can be adapted accordingly. It can assist in multi-sensor fusion calibration, improve the accuracy and reliability of target detection. The side flip camera 11 can provide richer perspective dimension data for the fusion system and assist in calibrating radar accuracy.

[0040] Please refer to the reference again. Figures 1 to 11 The working principle of the target detection system for autonomous sanitation vehicles provided by this invention is as follows: Step S1, target image detection; The drive motor 52 is started to drive the lead screw 53 to rotate. When the lead screw 53 rotates, it can control the vertical lifting and lowering of the moving plate 56. During the lifting and lowering process of the moving plate 56, the movement direction of the moving plate 56 can be restricted by the limit rod 54, thereby making the lifting and lowering movement of the moving plate 56 more stable. When the moving plate 56 drives the rotating gear 57 to move to the position of the fixed rack 51, the rotating gear 57 will mesh with the fixed rack 51 to drive the rotating plate 58 to rotate the image sensor 510 inside it. Thus, the lifting and rotation linkage of the moving plate 56 can be guaranteed during the lifting and lowering process. The image sensor 510 collects images through an RGB or infrared camera and combines image recognition technology to determine the target category and distinguish between "recyclable waste", "construction waste", "pedestrians" and "traffic signs". Step S2, side target image detection; The user can slide the mounting plate 4 horizontally. Whether the side plates 8 on both sides of the mounting plate 4 move to the left or the right, when the inclined surface on the side plate 8 moves to the position of the guide wheel 10, it will abut against the force control guide wheel 10, thereby realizing the entire rotating frame 9 to form a flipping motion in the horizontal direction. During the flipping process, the angle of the side camera 11 is switched, and the side camera 11 is switched from the horizontal visual angle to the flipping visual angle. In this way, image target detection can be performed on the side of the frame 1.

[0041] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made under the concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A target detection system for an autonomous sanitation vehicle, characterized in that, This includes the chassis that supports the hardware layer and the software layer; The software layer includes lidar, millimeter-wave radar, ultrasonic radar, and a data processing unit. The hardware layer includes a detection mechanism and a mounting frame. The detection mechanism includes a rotating plate and an image sensor. The image sensor acquires images through an RGB or infrared camera and combines image recognition technology to determine the target category, distinguishing between "recyclable waste," "construction waste," "pedestrians," and "traffic signs." LiDAR acquires the three-dimensional coordinates, contours, and distance information of targets by emitting laser point clouds, identifies the position and movement trajectory of garbage, low obstacles, pedestrians, and vehicles, and forms a three-dimensional image; The image sensor, lidar, millimeter-wave radar, and ultrasonic radar transmit the acquired image data to the data processing unit. The data processing unit integrates, processes, and transmits the various image data and performs basic target detection and recognition, including visual algorithms and lidar algorithms. Visual algorithms: Deep learning algorithms such as YOLO and Faster R-CNN are used to detect images. By training a "sanitation scene-specific dataset", the recognition rate of special targets is improved. LiDAR algorithm: Use algorithms such as PointNet and PointPillars to classify the clustered point cloud, identify targets such as "pedestrians, vehicles, garbage, and obstacles", and output three-dimensional image data such as the position, size and movement speed of the targets.

2. The target detection system for autonomous sanitation vehicles according to claim 1, characterized in that, The vehicle frame is bolted to a mounting bracket. Two sliding rods are installed inside the mounting bracket. An upper sliding frame and a lower sliding frame are slidably connected to the exterior of each sliding rod. A mounting plate is fixed between the upper and lower sliding frames. The detection mechanism also includes a first positioning plate and a second positioning plate fixed to the outer wall of the mounting plate. A drive motor is mounted outside the mounting plate and above the first positioning plate. A lead screw is keyway-connected to the output shaft of the drive motor. Limiting rods are provided on both sides of the lead screw outside the mounting plate. A fixed rack is bolted to the outer wall of the mounting plate and to one side of the lead screw. A movable plate is threaded to the outer wall of the lead screw. A rotating gear is rotatably connected to the outer wall of the movable plate. The rotating plate is bolted to the outer wall of the rotating gear. The image sensor is embedded inside the rotating plate. A linkage plate is bolted to the side wall of the movable plate.

3. The target detection system for autonomous sanitation vehicles according to claim 2, characterized in that, The upper and lower ends of the limiting rod are fixedly connected to the first positioning plate and the second positioning plate, and the two ends of the lead screw are rotatably connected to the first positioning plate and the second positioning plate.

4. The target detection system for autonomous sanitation vehicles according to claim 2, characterized in that, The movable plate and the limiting rod are slidably connected and can be raised and lowered along the vertical direction of the limiting rod. The rotating gear and the fixed rack mesh with each other.

5. The target detection system for autonomous sanitation vehicles according to claim 2, characterized in that, It also includes auxiliary mechanisms; The auxiliary mechanism includes a first limiting plate and a second limiting plate. The first limiting plate and the second limiting plate are respectively fixed to the outer wall of the mounting plate and located below the rotating plate. A rotating ring is rotatably connected inside the second limiting plate. A toothed ring is coaxially fixed to the side wall of the rotating ring. A cleaning roller is rotatably connected inside the first limiting plate. A ratchet is connected to the axis of the cleaning roller and located inside the rotating ring via a keyway. A positioning seat is fixed to the inner wall of the rotating ring. A ratchet is installed inside the positioning seat via a torsion spring.

6. The target detection system for autonomous sanitation vehicles according to claim 5, characterized in that, A movable rack is fixedly provided at the bottom of the linkage plate, and the movable rack and the toothed ring mesh with each other.

7. The target detection system for autonomous sanitation vehicles according to claim 5, characterized in that, The ratchet and the outer wall of the ratchet are meshed together. The swivel ring has a hollow design at its center and does not contact the ratchet.

8. The target detection system for autonomous sanitation vehicles according to claim 2, characterized in that, Side plates are bolted to both sides of the mounting plate. A rotating frame is rotatably mounted on the outer wall of the mounting frame via a torsion spring. A guide wheel is rotatably connected to the inner wall of the rotating frame. A side camera is mounted on the outer wall of the rotating frame in the same horizontal direction as the guide wheel.

9. The target detection system for autonomous sanitation vehicles according to claim 8, characterized in that, The width of the guide wheel is equal to the width of the side plate, and the side of the side plate has a sloping structure.

Images