Unmanned vehicle and sensing system

Abstract

The invention relates to an unmanned vehicle and a sensing system. The system comprises a camera assembly and a radar assembly. The camera assembly is arranged on the vehicle body and used for recognizing the environment around the vehicle; the radar assembly is arranged on the vehicle body and used for detecting obstacles around the vehicle; the radar assembly comprises a rear laser radar arranged at the tail of the vehicle, and the rear laser radar is used for recognizing rear obstacles and the trailer so as to achieve automatic butt joint of the trailer. The camera assembly and the radar assembly are connected with each other so as to achieve multi-layer environment perception redundancy and full-view coverage around the vehicle. The system has full-view coverage and multi-layer redundant environment perception capability, can significantly improve the recognition precision of rear obstacles and trailers, provides efficient and reliable technical support for automatic docking of trailers, and can significantly improve the operation efficiency and safety of unmanned vehicles in complex scenes.

Description

Technical Field

[0001] The present invention relates to the field of unmanned driving technology, and in particular to unmanned driving vehicles and perception systems. Background Art

[0002] In modern factory logistics and transportation, the efficient operation of automated logistics vehicles depends on accurate environmental perception capabilities. Traditional environmental perception systems usually use a single sensor to achieve target detection and positioning, but their performance has obvious limitations in complex scenarios. In addition, traditional technologies cannot match trailer systems, resulting in inefficient trailer docking and safety risks. Summary of the invention

[0003] Based on this, it is necessary to provide an unmanned vehicle and perception system to improve environmental perception capabilities and match trailers.

[0004] In a first aspect, an unmanned vehicle perception system is provided, the system comprising:

[0005] A camera assembly is arranged on the vehicle body and is used to identify the environment around the vehicle;

[0006] A radar component is arranged on the vehicle body and is used to detect obstacles around the vehicle;

[0007] The radar assembly includes a rear laser radar installed at the rear of the vehicle, which is used to identify obstacles and trailers behind to achieve automatic docking of the trailer;

[0008] The camera components and radar components are interconnected to achieve multi-layer environmental perception redundancy and full field of view coverage around the vehicle.

[0009] In one embodiment, the camera assembly includes:

[0010] The front camera group is arranged at the front of the vehicle to identify the lane lines, road signs and obstacles ahead;

[0011] Side-view camera groups are arranged on both sides of the vehicle body to identify adjacent lanes and lateral obstacles;

[0012] The rear-view camera is located at the rear of the vehicle and is used to identify obstacles behind it.

[0013] In one of the embodiments, the camera assembly is also used to fuse lane line curvature, road sign semantic information and obstacle contour data to achieve forward obstacle recognition.

[0014] In one embodiment, the front-view camera group includes at least two front-view cameras; and the side-view camera group includes at least four side-view cameras.

[0015] In one embodiment, the radar assembly further comprises:

[0016] The front millimeter-wave radar is placed at the front of the vehicle to detect the speed and distance information of obstacles ahead;

[0017] The side-view laser radar group is arranged on the front side of the vehicle body to detect obstacles in front and on the sides and generate a drivable area;

[0018] The top laser radar is arranged on the roof and is used for vehicle positioning and surround obstacle detection.

[0019] In one of the embodiments, the side-looking laser radar group includes at least two side-looking laser radars;

[0020] At least two side-looking lidars are used to detect the spatial position of obstacles in front and on the sides, and to generate the boundaries of the drivable area based on the images of the camera assembly.

[0021] In one embodiment, the radar assembly further comprises:

[0022] The ultrasonic radar group is installed around the vehicle body and is used for anti-collision detection around the vehicle.

[0023] In one embodiment, the system further comprises:

[0024] The vehicle computer device is used to fuse the environmental perception data detected by the front-view camera group with the environmental perception data detected by the side-view laser radar group to obtain first fused data, and generate a drivable area based on the first fused data; the vehicle computer device is also used to fuse the environmental perception data detected by the rear laser radar with the environmental perception data detected by the rear camera to obtain second fused data, and plan the trailer docking path based on the second fused data.

[0025] In one of the embodiments, the rear laser radar is also used to identify the spatial posture of the connection mechanism of the rear trailer and output the trailer docking control parameters to complete the automatic connection of the rear trailer.

[0026] In a second aspect, the present application provides an unmanned vehicle, which includes the unmanned vehicle perception system in any one of the embodiments of the first aspect.

[0027] The above-mentioned unmanned vehicle perception system achieves full-field coverage and multi-layer redundant environmental perception capabilities through the optimized layout and coordinated work of camera components and radar components; through the introduction of rear lidar, the recognition accuracy of rear obstacles and trailers is significantly improved, providing efficient and reliable technical support for automatic docking of trailers, which can significantly improve the operating efficiency and safety of unmanned vehicles in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A schematic diagram of the front view structure of an unmanned vehicle perception system according to an embodiment;

[0029] Figure 2 A schematic diagram of the side structure of an unmanned vehicle perception system according to another embodiment;

[0030] Figure 3 A schematic diagram of the field of view of a camera component and a radar component of an unmanned vehicle perception system according to an embodiment.

[0031] Reference numerals:

[0032] 111. Front-view camera; 112. Front-view camera; 121. Side-view camera; 122. Side-view camera; 123. Side-view camera; 124. Side-view camera; 130. Rear-view camera; 210. Front millimeter-wave radar; 221. Side-view lidar; 222. Side-view lidar; 230. Top lidar; 240. Rear lidar; 251. Ultrasonic radar; 252. Ultrasonic radar; 253. Ultrasonic radar; 254. Ultrasonic radar; 255. Ultrasonic radar; 256. Ultrasonic radar; 257. Ultrasonic radar; 258. Ultrasonic radar; 259. Ultrasonic radar; 2510. Ultrasonic radar; 2511. Ultrasonic radar; 2512. Ultrasonic radar. DETAILED DESCRIPTION

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0034] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0035] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0036] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0037] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0038] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "upper", "lower", "left", "right" and similar expressions used herein are for illustrative purposes only and are not intended to be the only implementation method.

[0039] In modern factory logistics and transportation, the efficient operation of automated logistics vehicles depends on accurate environmental perception capabilities. With the rapid development of industry and intelligent manufacturing, factory logistics scenarios have put forward higher requirements for the intelligence, safety and efficiency of automated logistics vehicles. However, traditional environmental perception systems still have many technical bottlenecks in practical applications and are difficult to meet the needs of complex scenarios.

[0040] Traditional environmental perception systems usually use a single sensor to achieve target detection and positioning. For example, a camera can provide rich visual information under normal lighting conditions, but its performance drops significantly under low light or strong light interference, resulting in reduced target recognition accuracy. On the other hand, although radar (such as millimeter wave radar) has strong anti-interference capabilities and all-weather working characteristics, it has limited perception capabilities for transparent or reflective objects and has difficulty providing high-precision target contour information. In addition, although ultrasonic radar is suitable for close-range detection, its detection range is limited and cannot meet the perception needs of medium and long distances.

[0041] In factory logistics scenarios, automated logistics vehicles not only need to sense environmental information in front and on the sides, but also need to accurately identify the trailer and the attachment position behind. However, the existing technology is not capable of identifying the trailer and the attachment position behind, resulting in inefficiency and safety hazards in the trailer docking process. For example, traditional methods usually rely on manual operation or simple visual inspection, which is not only time-consuming and labor-intensive, but also prone to errors in complex environments, making it difficult to meet the needs of efficient and accurate automation.

[0042] In response to the above problems, the present invention proposes an environmental perception system for factory logistics vehicles based on multi-sensor fusion. By optimizing the sensor layout scheme and fusion algorithm, it realizes the environmental perception capability of full field coverage and multi-layer redundancy. Its key technical points include:

[0043] System solution and sensor layout: Through the reasonable layout of cameras, laser radar, millimeter-wave radar and ultrasonic radar, the vehicle's front, side, rear and near-field environment are fully covered. Cameras are arranged at the front and sides of the vehicle to identify traffic signs, pedestrians and obstacles; laser radars are arranged at the top and rear of the vehicle to provide high-precision three-dimensional point cloud data; millimeter-wave radars are arranged at the four corners of the vehicle for medium and long-distance target detection; ultrasonic radars are arranged at the bottom and rear of the vehicle for close-range obstacle detection.

[0044] Recognize the rear trailer and the connection position through LiDAR: Use the high-precision three-dimensional point cloud data of LiDAR, combined with deep learning algorithms, to accurately identify the posture and connection position of the rear trailer, and support the automatic docking function of the trailer. Through point cloud segmentation, feature extraction and posture estimation technology, the system can calculate the relative position and angle of the trailer in real time, providing accurate guidance information for trailer docking.

[0045] The present invention significantly improves the reliability, accuracy and adaptability of the environmental perception system through the collaborative work of multiple sensors, and provides technical support for the safe and efficient operation of factory logistics vehicles. Compared with traditional methods, the advantages of the present invention are:

[0046] Full field of view coverage: Through the arrangement of multiple sensors, no blind spots in the vehicle's surrounding environment are achieved.

[0047] Multi-layer redundancy: Different sensors complement each other to ensure reliable environmental information in complex environments.

[0048] High-precision identification: The combination of LiDAR and deep learning algorithms enables accurate identification of the rear trailer and the attachment position.

[0049] Efficient docking: The automated trailer docking function significantly improves logistics and transportation efficiency and reduces the risk of manual operation.

[0050] In an exemplary embodiment, see Figure 1 and Figure 2 , provides an unmanned vehicle perception system, which includes a camera component and a radar component.

[0051] The camera assembly is arranged on the vehicle body and is used to identify the environment around the vehicle.

[0052] Optionally, the camera assembly is arranged on the vehicle body, mainly for identifying the environmental information around the vehicle. Exemplarily, the cameras are arranged on the front, side, top and rear of the vehicle to achieve full coverage perception of the road ahead, the lateral area, the top environment and the rear area. The camera can capture high-resolution image data, and combined with computer vision algorithms, it can identify targets such as traffic signs, pedestrians, vehicles, obstacles, etc. in real time. In addition, the camera can also provide rich texture and color information, providing important visual reference for the environmental perception system.

[0053] The radar component is installed on the vehicle body and is used to detect obstacles around the vehicle.

[0054] Optionally, the radar component is also arranged on the vehicle body to detect obstacles around the vehicle. Radar components include but are not limited to millimeter-wave radar, ultrasonic radar and laser radar, etc. Through the coordinated work of multiple types of radars, accurate detection of obstacles of different distances and characteristics can be achieved. For example, the millimeter-wave radar is arranged at the four corners of the vehicle, mainly used for medium and long-distance target detection, with strong anti-interference ability and all-weather working characteristics; the ultrasonic radar is arranged at the bottom and rear of the vehicle, suitable for close-range obstacle detection, especially in low-speed driving and parking scenarios.

[0055] The radar assembly includes a rear laser radar 240 disposed at the rear of the vehicle, and the rear laser radar 240 is used to identify rear obstacles and trailers to achieve automatic docking of the trailer.

[0056] Optionally, the radar assembly includes a rear laser radar 240 disposed at the rear of the vehicle for identifying rear obstacles and trailers. Exemplarily, the rear laser radar 240 generates high-precision three-dimensional point cloud data by emitting a laser beam and receiving a reflected signal, and can accurately identify the position, shape and posture of rear obstacles. Combined with a deep learning algorithm, the rear laser radar 240 can further identify the coupling position and posture of the rear trailer, and provide accurate guidance information for automatic docking of the trailer. Through point cloud segmentation, feature extraction and posture estimation technology, the system can calculate the relative position and angle of the trailer in real time to ensure the efficiency and safety of the trailer docking process.

[0057] The camera components and radar components are interconnected to achieve multi-layer environmental perception redundancy and full field of view coverage around the vehicle.

[0058] Optionally, the camera component and the radar component are interconnected to achieve multi-layer environmental perception redundancy and full field of view coverage around the vehicle through a data fusion algorithm. For example, the visual information provided by the camera and the distance information provided by the radar complement each other to ensure reliable environmental perception capabilities in complex environments (such as low light, rainy and foggy weather, or dynamic obstacle interference). For example, in low light conditions, the camera may not be able to accurately identify the target, but the radar can still provide distance information of the obstacle; and when the radar has limited perception of transparent or reflective objects, the camera can provide supplementary visual information. Through the collaborative work of multiple sensors, the system can achieve no-dead-angle perception of the vehicle's surroundings, providing comprehensive protection for the safe operation of unmanned vehicles.

[0059] The above-mentioned unmanned vehicle perception system achieves full-view coverage and multi-layer redundant environmental perception capabilities through the optimized arrangement and coordinated work of camera components and radar components; through the introduction of the rear laser radar 240, the recognition accuracy of rear obstacles and trailers is significantly improved, providing efficient and reliable technical support for automatic trailer docking, and significantly improving the operating efficiency and safety of unmanned vehicles in complex scenarios.

[0060] In an exemplary embodiment, see Figure 1 and Figure 2 The camera assembly includes a front camera group, a side view camera group and a rear view camera 130.

[0061] The front-view camera group is arranged at the front of the vehicle and is used to identify the lane lines, road signs and obstacles ahead.

[0062] The side-view camera group is arranged on both sides of the vehicle body and is used to identify adjacent lanes and lateral obstacles.

[0063] The rear-view camera 130 is arranged at the rear of the vehicle and is used to identify obstacles behind.

[0064] Exemplarily, the front-view camera group is arranged at the front of the vehicle, mainly used to identify the lane line ahead, the road signs ahead and the obstacles ahead. The front-view camera group includes at least two high-resolution cameras, one for wide-angle field of view coverage and the other for long-distance target recognition. The wide-angle camera can capture environmental information in a large range in front of the vehicle, and is suitable for identifying lane lines, traffic signs and close-range obstacles; while the long-range camera focuses on identifying long-distance targets on the road ahead, such as vehicles ahead, pedestrians or other obstacles. Through binocular vision technology, the front-view camera group can generate depth information of the front environment and provide the vehicle with accurate distance perception capabilities. In addition, the front-view camera group can also identify key information such as traffic lights and speed limit signs on the road ahead, providing important reference for the vehicle's decision-making system.

[0065] The side-view camera group is arranged on both sides of the vehicle body, mainly used to identify adjacent lanes and lateral obstacles. The side-view camera group includes one or more cameras on the left and right sides, covering the areas on both sides of the vehicle respectively. The side-view camera can monitor the vehicle dynamics in the adjacent lanes in real time, identify lateral obstacles (such as pedestrians, bicycles or other vehicles), and provide key environmental perception data for vehicle operations such as lane changing and overtaking. In addition, the side-view camera group can also identify lateral road signs (such as stop lines, turn signs, etc.) to provide support for vehicle path planning and decision-making. Through the data from the side-view camera group, the system can effectively avoid the safety hazards caused by blind spots and ensure the safe operation of the vehicle in complex traffic environments.

[0066] The rear-view camera 130 is arranged at the rear of the vehicle and is mainly used to identify obstacles behind the vehicle. The rear-view camera 130 uses a high-resolution sensor to capture the environmental information behind the vehicle in real time, including the rear vehicles, pedestrians or other obstacles. Combined with the image processing algorithm, the rear-view camera 130 can identify the movement trajectory and relative distance of the rear target, and provide accurate guidance information for the vehicle's reversing, parking and other operations. In addition, the rear-view camera 130 can also work in conjunction with radar components (such as rear lidar) to provide multi-level perception redundancy, further improving the reliability and accuracy of rear environment perception.

[0067] In the above embodiment, the camera assembly achieves all-round perception of the vehicle's surrounding environment through the optimized layout and functional integration of the front-view camera group, the side-view camera group and the rear-view camera 130, providing accurate and reliable data support for the path planning, obstacle recognition and decision-making control of the unmanned vehicle.

[0068] In an exemplary embodiment, see Figure 1 and Figure 2 The front-view camera group includes at least two front-view cameras; the side-view camera group includes at least four side-view cameras.

[0069] Exemplarily, the front view camera group includes a front view camera 111 and a front view camera 112 symmetrically distributed at the front end of the vehicle body; the side view cameras include a side view camera 121, a side view camera 122, a side view camera 123 and a side view camera 124 symmetrically distributed on both sides of the vehicle body.

[0070] In an exemplary embodiment, the camera assembly is also used to fuse lane line curvature, road sign semantic information and obstacle contour data to achieve forward obstacle recognition.

[0071] For example, the camera component can accurately identify the curvature information of the lane line ahead through high-resolution image acquisition and deep learning algorithms; the camera component can also identify and extract the semantic information of road signs, including but not limited to speed limit signs, stop signs, turn indicators, traffic lights, etc.; through image recognition algorithms, the system can extract the type, location and status of road signs from the images collected by the camera; the camera component can accurately identify the contour data of the obstacles ahead through image segmentation and contour extraction algorithms. The camera component uses multi-sensor data fusion algorithms (such as Kalman filtering, Bayesian networks, etc.) to comprehensively analyze the lane line curvature, road sign semantic information and obstacle contour data collected by the camera with the real-time status of the vehicle (such as vehicle speed, acceleration, steering angle, etc.), and generate a complete perception model of the environment ahead to achieve more accurate obstacle recognition ahead.

[0072] In an exemplary embodiment, see Figure 1 and Figure 2 The radar assembly also includes a front millimeter-wave radar 210, a side-looking lidar group and a top lidar 230.

[0073] The front millimeter-wave radar 210 is arranged at the front of the vehicle and is used to detect the speed and distance information of the obstacles ahead.

[0074] The side-view laser radar group is arranged on the front side of the vehicle body and is used to detect obstacles in front and on the sides and generate a drivable area.

[0075] The top laser radar 230 is arranged on the roof and is used for vehicle positioning and surround obstacle detection.

[0076] Exemplarily, the front millimeter wave radar 210 is arranged at the front of the vehicle, and is mainly used to detect the speed and distance information of the obstacle ahead. Its working principle is based on the transmission and reception of millimeter wave signals, and by measuring the time difference and Doppler shift of the reflected signal, the distance and relative speed of the obstacle ahead are calculated. The millimeter wave radar has all-weather performance, high-precision ranging, real-time speed detection and other functions to achieve adaptive cruise control, emergency brake assistance and front obstacle warning.

[0077] The side-view laser radar group is arranged on the front side of the vehicle body, mainly used to detect obstacles in front and on the sides, and generate information about the drivable area. The laser radar generates high-precision three-dimensional point cloud data by emitting laser beams and receiving reflected signals, and can accurately identify the shape, position and size of obstacles. The side-view laser radar group has a large field of view, which can cover the front and side areas of the vehicle, providing comprehensive environmental perception. By analyzing the point cloud data, the system can identify the drivable area and provide a basis for the vehicle's path planning.

[0078] The top LiDAR 230 is arranged on the roof and is mainly used for vehicle positioning and surround obstacle detection. Its working principle is similar to that of the side-view LiDAR group, but it has a higher installation position and a wider coverage range. By combining high-precision maps and LiDAR data, centimeter-level positioning of the vehicle can be achieved to ensure accurate navigation of the vehicle in complex environments. The top LiDAR 230 can cover a 360° range around the vehicle, providing comprehensive environmental perception capabilities. By analyzing point cloud data and generating a three-dimensional environmental model around the vehicle in real time, it can identify obstacles around the vehicle (such as pedestrians, vehicles, buildings, etc.) and generate avoidance strategies, providing comprehensive data support for path planning and decision-making.

[0079] The radar assembly in this embodiment achieves all-round perception and obstacle detection of the vehicle's surrounding environment through the coordinated work of the front millimeter-wave radar 210, the side-looking laser radar group and the top laser radar 230, ensuring the safe operation of the unmanned vehicle.

[0080] In an exemplary embodiment, see Figure 1 and Figure 2 , the side-looking laser radar group includes at least two side-looking laser radars.

[0081] At least two side-looking lidars are used to detect the spatial position of obstacles in front and on the sides, and to generate the boundaries of the drivable area based on the images of the camera assembly.

[0082] Exemplarily, the side-view laser radar group includes a side-view laser radar 221 and a side-view laser radar 222 symmetrically distributed on both sides of the vehicle body. The side-view laser radar group and the camera component work together through data fusion algorithms (such as deep learning, Kalman filtering, etc.) to provide all-round perception of the vehicle's surrounding environment. The laser radar provides high-precision point cloud data, and the camera provides rich visual information. The two generate a more accurate environmental perception model through data fusion algorithms. The camera component identifies the features of the drivable area such as lane lines and shoulders, and the laser radar provides the spatial position information of obstacles. The two are combined to generate an accurate boundary of the drivable area. At the same time, the camera component identifies the category of obstacles, and the laser radar provides the spatial position information of obstacles. The two are combined to generate a complete obstacle perception model.

[0083] In the above embodiment, the side-view laser radar group realizes accurate detection of obstacles in front of and to the side of the vehicle by deploying at least two side-view laser radars, and generates the boundary of the drivable area in combination with the image data of the camera assembly. The fusion of its high-precision point cloud data and visual information provides the vehicle with comprehensive environmental perception capabilities, ensuring safe driving in complex traffic environments.

[0084] In an exemplary embodiment, see Figure 1 and Figure 2 , the radar component also includes an ultrasonic radar group.

[0085] The ultrasonic radar group is installed around the vehicle body and is used for anti-collision detection around the vehicle.

[0086] Exemplarily, the ultrasonic radar group includes 12 ultrasonic radars, including ultrasonic radar 251 and ultrasonic radar 252 symmetrically distributed on the front of the vehicle body, ultrasonic radar 253 and ultrasonic radar 254 distributed on the rear of the vehicle body, and ultrasonic radar 255, ultrasonic radar 256, ultrasonic radar 257, ultrasonic radar 258, ultrasonic radar 259, ultrasonic radar 2510, ultrasonic radar 2511, and ultrasonic radar 2512 symmetrically distributed on both sides of the vehicle body. Ultrasonic radar can accurately measure the distance between the vehicle and surrounding obstacles by emitting ultrasonic signals and receiving reflected signals, and provide real-time and reliable anti-collision warning information. Its working principle is based on the emission and reception of ultrasonic waves. By measuring the time difference from the emission to the reflection of ultrasonic signals, the distance between the vehicle and the obstacle is calculated to achieve parking assistance, low-speed collision avoidance, blind spot detection and other functions.

[0087] In an emergency, the ultrasonic radar group can be used as a redundant system to provide additional safety for the vehicle. The ultrasonic radar group works in conjunction with other sensors (such as millimeter-wave radar and lidar). When a sensor fails or the detection result is uncertain, the ultrasonic radar can provide supplementary information to ensure the reliability of environmental perception. When the system detects that the distance between the vehicle and the obstacle is too close, the ultrasonic radar can trigger emergency braking to avoid a collision. In complex environments, ultrasonic radar can provide accurate short-range detection information to make up for the shortcomings of other sensors.

[0088] In an exemplary embodiment, the system also includes a vehicle computing device.

[0089] The vehicle computer device is used to fuse the environmental perception data detected by the front-view camera group with the environmental perception data detected by the side-view laser radar group to obtain first fused data, and generate a drivable area according to the first fused data.

[0090] The vehicle computer device is also used to fuse the environmental perception data detected by the rear laser radar with the environmental perception data detected by the rear camera to obtain second fused data, and perform trailer docking path planning based on the second fused data.

[0091] For example, the front camera group captures the environmental perception data in front of the vehicle and identifies the features of the drivable area such as lane lines, shoulders, and traffic signs; the side-view laser radar group generates high-precision three-dimensional point cloud data to detect the spatial position of obstacles in front and on the sides; the first fusion data is obtained by fusing the environmental perception data of the front camera group with the point cloud data of the side-view laser radar group, and the vehicle computer device generates an accurate boundary of the drivable area based on the first fusion data. The system can dynamically update the drivable area based on real-time perception data to ensure the safe driving of the vehicle in complex environments.

[0092] The rear laser radar 240 generates three-dimensional point cloud data behind the vehicle to detect the position, posture and surrounding obstacles of the trailer; the rear camera captures the visual information in front of the vehicle to identify the position and posture of the trailer docking target; the second fusion data is obtained by fusing the point cloud data of the rear laser radar 240 with the visual information of the rear camera, and the vehicle computer device generates the precise path for the trailer docking based on the second fusion data. For example, the laser radar detects the spatial position of the trailer, and the camera identifies the docking target of the trailer. The two are combined to plan the optimal docking path for the vehicle computer device.

[0093] In an exemplary embodiment, the rear laser radar 240 is also used to identify the spatial posture of the connection mechanism of the rear trailer and output trailer docking control parameters to complete the automatic connection of the rear trailer.

[0094] Exemplarily, the rear laser radar 240 emits a laser beam and receives a reflected signal to generate three-dimensional point cloud data of the trailer connection mechanism and its surroundings; through intelligent algorithms (such as deep learning, point cloud segmentation, etc.), the system extracts the features of the trailer connection mechanism (such as position, angle, size, etc.) from the point cloud data; based on the extracted features, the system calculates the spatial posture of the trailer connection mechanism (such as horizontal position, vertical height, tilt angle, etc.). After identifying the spatial posture of the trailer connection mechanism, the system plans the docking path of the vehicle based on the posture information of the trailer connection mechanism to ensure that the vehicle and the trailer connection mechanism are aligned; the system generates control parameters for trailer docking (such as steering wheel angle, vehicle speed, braking distance, etc.) and transmits these parameters to the vehicle's control system. During the docking process, the system dynamically adjusts the control parameters based on real-time perception data to ensure the accuracy and safety of docking.

[0095] In this embodiment, the rear laser radar 240 realizes the spatial posture recognition and automatic docking control of the trailer connection mechanism through high-precision three-dimensional point cloud data and intelligent algorithms. Its high-precision perception, real-time performance and adaptability to complex environments provide comprehensive technical support for trailer docking, significantly improving the automation level and safety of trailer docking.

[0096] In an exemplary embodiment, the field of view of the camera assembly and the radar assembly is as follows: Figure 3 As shown in the figure, the data fusion algorithm realizes multi-layer environmental perception redundancy and full field of view coverage around the vehicle, achieving no blind spot perception of the vehicle's surrounding environment and providing comprehensive protection for the safe operation of unmanned vehicles.

[0097] In an exemplary embodiment, an unmanned vehicle is provided, which includes the unmanned vehicle perception system described in any one of the above embodiments.

[0098] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An unmanned vehicle perception system, characterized in that: include: A camera assembly is arranged on the vehicle body and is used to identify the environment around the vehicle; A radar component is installed on the vehicle body and is used to detect obstacles around the vehicle; The radar assembly includes a rear laser radar disposed at the rear of the vehicle, and the rear laser radar is used to identify rear obstacles and trailers to achieve automatic docking of the trailers; The camera assembly and the radar assembly are interconnected to achieve multi-layer environmental perception redundancy and full field of view coverage around the vehicle.

2. The unmanned vehicle perception system according to claim 1, characterized in that: The camera assembly comprises: The front camera group is arranged at the front of the vehicle to identify the lane lines, road signs and obstacles ahead; Side-view camera groups are arranged on both sides of the vehicle body to identify adjacent lanes and lateral obstacles; The rear-view camera is located at the rear of the vehicle and is used to identify obstacles behind it.

3. The unmanned vehicle perception system according to claim 2, characterized in that: The camera assembly is also used to fuse lane line curvature, road sign semantic information and obstacle contour data to achieve forward obstacle recognition.

4. The unmanned vehicle perception system according to claim 2, characterized in that: The front-view camera group includes at least two front-view cameras; the side-view camera group includes at least four side-view cameras.

5. The unmanned vehicle perception system according to claim 1, characterized in that: The radar assembly further comprises: The front millimeter-wave radar is placed at the front of the vehicle to detect the speed and distance information of obstacles ahead; The side-view laser radar group is arranged on the front side of the vehicle body to detect obstacles in front and on the sides and generate a drivable area; The top laser radar is arranged on the roof and is used for positioning the vehicle and detecting obstacles around it.

6. The unmanned vehicle perception system according to claim 5, characterized in that: The side-looking laser radar group includes at least two side-looking laser radars; The at least two side-looking laser radars are used to detect the spatial positions of front and side obstacles and generate the boundaries of the drivable area based on the images of the camera assembly.

7. The unmanned vehicle perception system according to claim 5, characterized in that: The radar assembly further comprises: The ultrasonic radar group is arranged around the vehicle body and is used for anti-collision detection around the vehicle.

8. The unmanned vehicle perception system according to claim 5, characterized in that: The system further comprises: The vehicle computer device is used to fuse the environmental perception data detected by the front-view camera group with the environmental perception data detected by the side-view laser radar group to obtain first fusion data, and generate the drivable area according to the first fusion data; the vehicle computer device is also used to fuse the environmental perception data detected by the rear laser radar with the environmental perception data detected by the rear camera to obtain second fusion data, and plan the trailer docking path according to the second fusion data.

9. The unmanned vehicle perception system according to claim 1, characterized in that: The rear laser radar is also used to identify the spatial posture of the connection mechanism of the rear trailer and output trailer docking control parameters to complete the automatic connection of the rear trailer.

10. An unmanned vehicle, characterized in that: The unmanned vehicle includes an unmanned vehicle perception system as described in any one of claims 1-9.

Images