High-speed road sweeper, obstacle avoidance method and system, medium and computer program product

Abstract

The invention discloses a high-speed road sweeper, an obstacle avoidance method and system, a medium and a computer program product, and the method comprises the steps: obtaining the detection result of each sensor when a controller detects that a plurality of sensors detect the same obstacle at the same time; the controller dynamically adjusts the confidence coefficient weight of each sensor based on the real-time environment information and the vehicle operation mode, determines a reward and punishment factor according to the detection result of each sensor, and calculates the comprehensive confidence coefficient according to the confidence coefficient weight and the reward and punishment factor. The controller is configured to store a threshold interval and send a corresponding decision instruction to the high-speed road sweeper according to the threshold interval where the comprehensive confidence is located; according to the invention, when the detection result of the sensor conflicts, the confidence is dynamically adjusted according to the real-time environment information and the vehicle operation mode so as to determine the comprehensive confidence, and when the comprehensive confidence of the target exceeds the preset threshold, the corresponding obstacle avoidance control instruction is generated, so that accurate obstacle avoidance in a complex scene can be realized.

Description

Technical Field

[0001] This invention relates to a method, system, medium, and computer program product for obstacle avoidance of a high-speed road sweeper, belonging to the field of intelligent assisted driving technology for special vehicles. Background Technology

[0002] Traffic flow on highways is generally between 60 and 120 km / h. Highway sweepers need to operate in the flow of traffic. Drivers need to pay attention to road conditions while maintaining a reasonable distance from the road edge at all times. This requires high driving skills and concentration. Using an intelligent obstacle avoidance system can effectively reduce the burden on drivers.

[0003] The following problems exist in the related technologies: Pure radar systems (such as ultrasonic or millimeter-wave radar) can only detect the presence and distance of obstacles, but cannot identify the type of obstacle, resulting in a high false alarm rate. False signals can negatively affect the driver's safe operation. When facing a turn, the distance indicated by the lateral radar will not be able to provide accurate information about the road conditions ahead, causing the detection of some road sections to fail.

[0004] It is evident that, in order to solve the above-mentioned technical problems, there is an urgent need for a high-speed road sweeper, obstacle avoidance method, system, medium, and computer program product. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a high-speed road sweeper, obstacle avoidance method, system, medium and computer program product that can dynamically adjust the confidence level by combining real-time environmental information and vehicle operation mode to achieve accurate obstacle avoidance in complex scenarios.

[0006] To achieve the above objectives / to solve the above technical problems, the present invention is implemented using the following technical solution: In a first aspect, the present invention provides a method for obstacle avoidance by a high-speed road sweeper, comprising: When the controller detects that multiple sensors simultaneously detect the same obstacle, it acquires the detection results from each sensor. The controller dynamically adjusts the confidence weights of each sensor based on real-time environmental information and vehicle operating mode, and determines the reward and penalty factors based on the detection results of each sensor, so as to calculate the comprehensive confidence level based on the confidence weights and reward and penalty factors. The controller is configured to store threshold intervals and send corresponding decision instructions to the high-speed sweeper based on the threshold interval in which the comprehensive confidence level falls.

[0007] Furthermore, the real-time environmental information includes one or more of weather conditions, lighting conditions, and vehicle speed.

[0008] Furthermore, when the controller detects that multiple sensors simultaneously detect the same obstacle, the detection results of each sensor are preceded by: In response to a work instruction, the controller acquires vehicle operating status data and environmental perception data from multiple sensors; The controller identifies curves based on operating status data and environmental perception data; The controller determines the operating mode of the high-speed sweeper based on the operating status data and curve recognition results.

[0009] Furthermore, the environmental perception data includes point cloud data, image data, and distance data, and the vehicle operating status data includes steering wheel angle and vehicle speed; The controller performs curve recognition based on operating status data and environmental perception data, including: The road curvature is determined based on the point cloud data or the image data, and the road boundary is fitted to the point cloud data. When the road curvature is greater than a first preset threshold, the steering wheel angle is greater than a second preset threshold within a preset duration, and the road boundary fitting result is a curve, it is determined to be a curve.

[0010] Furthermore, the controller determines the operating mode of the high-speed sweeper truck based on the operating status data and curve recognition results, including: Based on the current vehicle speed and curve recognition results, the operating mode is determined to be one of the following three: If the current vehicle speed is lower than the third preset threshold, it is determined to be a low-speed edge-keeping operation mode; If the current vehicle speed is higher than the third preset threshold and the curve recognition result is a curve, then it is determined to be a high-speed curve operation mode. If the current vehicle speed is higher than the third preset threshold and the curve recognition result is not a curve, then it is determined to be a straight road high-speed operation mode.

[0011] Furthermore, in the high-speed curve operation mode, the dynamic adjustment of the confidence weights of each sensor includes: Reduce the corresponding weights of sensors inside the vehicle.

[0012] Furthermore, the formula for calculating the overall confidence level is as follows: ; In the formula, To calculate the overall confidence level, n represents the number of sensors that detected the same target; ci represents the independent confidence level of the i-th sensor for that target; wi represents the dynamic weighting factor of the i-th sensor in the current environment; m represents the number of applicable reward / penalty factors; and bj represents the j-th reward / penalty factor. ; In the formula, The base weights for the i-th sensor are: This is a weather adjustment factor. This is the illumination adjustment factor. This is the vehicle speed adjustment coefficient. Adjustment coefficient for vehicle operating mode; The determination of reward and penalty factors based on the detection results of each sensor includes: A reward value is assigned when the detection results of all sensors are consistent, and a penalty value is assigned when the detection results of all sensors conflict. The formula for calculating the reward and punishment factor is as follows: ; In the formula, Rewards are matched to the type. For consistent speed, Rewards are given for consistency in trajectory continuity. Penalty for sensor collisions.

[0013] Secondly, the present invention provides a high-speed road sweeper, comprising: Vehicle body; The sensing unit includes multiple sensors for acquiring point cloud data, image data, and distance data of the front and sides of the vehicle body; A controller, which is signal-connected to the sensing unit, is configured to execute the obstacle avoidance method for the high-speed sweeper as described in the first aspect.

[0014] Thirdly, the present invention provides a high-speed road sweeper obstacle avoidance system, comprising: a controller; The controller is configured to include: The information acquisition module is configured to acquire the detection results of each sensor when multiple sensors detect the same obstacle simultaneously. The calculation module is configured to dynamically adjust the confidence weights of each sensor based on real-time environmental information and vehicle operating mode, and determine the reward and penalty factors based on the detection results of each sensor, so as to calculate the comprehensive confidence based on the confidence weights and reward and penalty factors. The judgment module is configured to store threshold intervals and send corresponding decision instructions to the high-speed sweeper based on the threshold interval in which the comprehensive confidence level falls.

[0015] Fourthly, the present invention provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the obstacle avoidance method for high-speed sweeper trucks described in the first aspect.

[0016] Fifthly, the present invention provides a non-transitory computer-readable storage medium storing computer device readable instructions that, when executed by at least one processor, cause the steps of the high-speed sweeper obstacle avoidance method as described in the first aspect to be performed.

[0017] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: When sensor detection results conflict, this invention dynamically adjusts the confidence level based on real-time environmental information and vehicle operating mode to determine the overall confidence level. When the overall confidence level of the target exceeds a preset threshold, a corresponding obstacle avoidance control command is generated, enabling precise obstacle avoidance in complex scenarios. Attached Figure Description

[0018] Figure 1 This is a flowchart of the obstacle avoidance method for high-speed road sweepers provided by the present invention; Figure 2 A schematic diagram of the high-speed sweeper provided by the present invention; Figure 3 This is a schematic diagram of the hardware architecture of the high-speed sweeper provided by the present invention; Figure 4 A flowchart illustrating the steps of dynamic confidence adjustment in the obstacle avoidance method for high-speed sweeper trucks provided by this invention; Figure 5 This is a flowchart illustrating the obstacle avoidance method for high-speed road sweepers provided by the present invention; Figure 6 This is a flowchart illustrating the confidence fusion calculation and decision-making process in the obstacle avoidance method for high-speed road sweepers provided by this invention. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0020] In the description of this invention, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0021] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0022] like Figure 6 As shown, in one embodiment of the present invention, the obstacle avoidance method for a high-speed sweeper includes: When the controller detects that multiple sensors simultaneously detect the same obstacle, it acquires the detection results from each sensor. The controller dynamically adjusts the confidence weights of each sensor based on real-time environmental information and vehicle operating mode, and determines the reward and penalty factors based on the detection results of each sensor, so as to calculate the comprehensive confidence level based on the confidence weights and reward and penalty factors. The controller is configured to store threshold intervals and send corresponding decision instructions to the high-speed sweeper based on the threshold interval in which the comprehensive confidence level falls.

[0023] In the above technical solutions, such as Figure 4 and Figure 6 As shown, each sensor's detection result is assigned a basic confidence level. When multiple sensors detect the same target, their confidence levels are weighted and superimposed. During the weighted superposition, when sensor detection results conflict, the confidence level is dynamically adjusted according to the sensor type, real-time environmental information, and vehicle operating mode to determine the comprehensive confidence level. When the comprehensive confidence level of the target exceeds the preset threshold, a corresponding obstacle avoidance control command is generated. like Figure 6 As shown, the preset thresholds include a warning threshold, a preparation threshold, and an execution threshold. Based on these thresholds, four threshold intervals are defined: the first interval is below the warning threshold; the second interval is above or equal to the warning threshold but below the preparation threshold; the third interval is above or equal to the preparation threshold but below the execution threshold; and the fourth interval is above or equal to the execution threshold. When the overall confidence level is in the first interval, the target is ignored and considered as noise or a false alarm. When the overall confidence level is in the second range, the driver warning state is activated, and an audible and visual alarm is issued to the driver, but control is not intervened at this time. When the overall confidence level is in the third interval, the system enters the obstacle avoidance preparation state, ready to actively brake or steer at any time. When the overall confidence level is in the fourth interval, active obstacle avoidance is performed.

[0024] The formula for calculating the overall confidence level is as follows: ; in: n: The number of sensors that detected the target; ci: The independent confidence level of the i-th sensor for the target; wi: The dynamic weighting factor of the i-th sensor in the current environment (0≤wi≤1.5, default value is 1); m: The number of applicable reward / penalty factors; bj: The j-th reward / penalty factor, used to add points to specific high-value information or deduct points from contradictory information, as detailed in Table 2 below; The dynamic weighting factor wi is dynamically adjusted according to the environmental context, and its calculation is based on the product of multiple adjustment coefficients: ; The initial parameter allocation is shown in Table 1.

[0025] Table 1

[0026] The reward / penalty factor bj is used to handle cooperation and conflict between sensors, and its calculation is based on a combination of multiple reward and penalty values: ; The parameter allocation is shown in Table 2.

[0027] Table 2

[0028] The system compares the calculated overall threat confidence score (Ctotal) with multiple preset thresholds to trigger different levels of response, as shown below: ; Threshold example: The warning threshold Twarn = 0.3; Set the threshold Tprepare to 0.5; The execution threshold Texecute=0.8.

[0029] In one embodiment of the present invention, the real-time environmental information includes one or more of weather conditions, lighting conditions, and vehicle speed.

[0030] In one embodiment of the present invention, such as Figure 5 As shown, when multiple sensors simultaneously detect the same obstacle via the controller, the detection results of each sensor are preceded by: In response to a work instruction, the controller acquires vehicle operating status data and environmental perception data from multiple sensors; The controller identifies curves based on operating status data and environmental perception data; The controller determines the operating mode of the high-speed sweeper based on the operating status data and curve recognition results.

[0031] In one embodiment of the present invention, the environmental perception data includes point cloud data from millimeter-wave radar, image data from a vision camera, and distance data from ultrasonic radar, and the vehicle operating status data is obtained based on vehicle CAN bus data, including steering wheel angle and vehicle speed. The controller performs curve recognition based on operating status data and environmental perception data, including: The road curvature is determined based on the point cloud data or the image data, and the road boundary is fitted to the point cloud data. When the road curvature is greater than a first preset threshold, the steering wheel angle is greater than a second preset threshold within a preset duration, and the road boundary fitting result is a curve, it is determined to be a curve.

[0032] Specifically, the controller projects the point cloud onto the image from the received forward radar and uses a calibrated extrinsic matrix to achieve target association. A visual algorithm identifies the types of obstacles in the image, while a lane detection algorithm calculates the current lane curvature. By combining these multiple pieces of information, the controller performs curve recognition based on the lane conditions. ① Visual judgment: The road curvature is calculated using a lane line detection algorithm. If the curvature is greater than a threshold, it is initially determined to be a curve. ② Vehicle signal judgment: If the steering wheel turns more than the threshold continuously, it is initially judged as a curve; ③ Radar point cloud judgment: perform boundary fitting on the front radar point cloud. If the boundary is curved, it is initially judged to be a curve. If all of the above conditions are met, the road condition can be determined to be a curve.

[0033] In one embodiment of the present invention, the controller determines the operating mode of the high-speed sweeper based on operating status data and curve recognition results, including: like Figure 5 As shown, based on the current vehicle speed and curve recognition results, the operating mode is determined to be one of the following three: If the current vehicle speed is lower than the third preset threshold (in this embodiment, it is...) Figure 5 If the speed is 20km / h, it is determined to be a low-speed edge-following operation mode. In this mode, the decision is mainly based on the lateral ultrasonic radar to achieve edge-following control with centimeter-level precision. The front sensor is responsible for safety and prevents collisions with obstacles that suddenly appear in front. If the current vehicle speed is higher than the third preset threshold and the curve recognition result is a curve, it is determined to be a high-speed curve operation mode. In this mode, the system switches to curve obstacle avoidance logic, suppresses or significantly relaxes the alarm thresholds of lateral ultrasonic and corner radar, because it is normal to approach the inside of the curve at this time. At the same time, the control core switches to path tracking, uses the road boundary perceived by the front vision and radar to generate a desired path, and determines whether the vehicle deviates from the path based on the path. If the current vehicle speed is higher than the third preset threshold and the curve recognition result is not a curve, it is determined to be a straight road high-speed operation mode. In this mode, the decision is mainly based on the fusion result of the front imaging millimeter-wave radar and vision, with the side sensor as an auxiliary.

[0034] In one embodiment of the present invention, in the high-speed operation mode on curves, the dynamic adjustment of the confidence weights of each sensor includes: Reduce the corresponding weights of sensors inside the vehicle.

[0035] like Figure 2 and Figure 3 As shown, in one embodiment of the present invention, a high-speed road sweeper is also provided, comprising: Vehicle body; The sensing unit includes multiple sensors for acquiring point cloud data, image data, and distance data of the front and sides of the vehicle body; A controller, which is signal-connected to the sensing unit, is configured to execute the obstacle avoidance method for the high-speed sweeper as described in the first aspect.

[0036] And the execution layer, which includes the vehicle's drive-by-wire system, braking system, and steering system.

[0037] Multiple sensors include a front long-range perception module and a side and blind spot perception module. The front long-range perception module includes at least one 4D imaging millimeter-wave radar and at least one vision camera, which are responsible for detecting long-range obstacles in front of the vehicle, measuring speed, and providing sparse point cloud and semantic information.

[0038] The side and blind spot perception module includes millimeter-wave angular radars installed at the left and right front corners of the vehicle, as well as ultrasonic radars distributed around the vehicle body, which are responsible for covering the oblique front blind spot and detecting near-field low-speed obstacles. refer to Figure 3 The system hardware connection is as follows: the 4D imaging millimeter-wave radar, vision camera, corner radar, and ultrasonic radar are all connected to the central processor (i.e., controller) via CAN bus or high-speed Ethernet. The central processor decision control module then communicates with the drive-by-wire chassis execution module via the vehicle CAN bus.

[0039] The 4D imaging millimeter-wave radar uses Arbe Robbin series processing chips and has a detection range of 250-350 meters, which meets the warning distance of 200 meters in advance for high-speed vehicles. The speed resolution is ≤0.1m / s, which can meet the requirements for the identification of moving and stationary obstacles. The vision camera is a global shutter binocular stereo camera, which can effectively avoid the jelly effect when the vehicle is traveling at high speed. The corner radar uses a 76~81GHz short-range radar with a detection range of 60~70 meters and a horizontal field of view of ≥150°, achieving seamless coverage of the sides of the vehicle. The ultrasonic radar is a digital ultrasonic radar with a detection range of 0.15 to 5 meters, strong anti-interference ability, and coverage of close-range working conditions. The central processing unit uses a domain controller based on the NVIDIA Jetson AGX Orin platform, with a computing power of ≥50tps; A base confidence level is assigned to each sensor: 0.7 for 4D millimeter-wave radar, 0.6 for vision camera, 0.6 for corner radar, and 0.3 for ultrasonic radar. These confidence levels are not static but dynamically adjusted based on context. When multiple sensors detect the same target simultaneously, their confidence levels are weighted and summed to obtain a comprehensive confidence level. When sensor findings conflict, the system performs intelligent arbitration and dynamically adjusts the weights. The system dynamically adjusts the confidence weights of each sensor based on real-time environmental information (such as weather, lighting, vehicle speed, and vehicle operating mode). For example, in heavy rain, the weight of the vision sensor is reduced, while the weight of the millimeter-wave radar is increased; in curve mode, the weight of the ultrasonic radar inside the vehicle is significantly reduced.

[0040] The following section will introduce the obstacle avoidance methods used by the aforementioned high-speed sweeper trucks, using specific application examples shown in Tables 3 and 4 as examples for data calculation.

[0041] Table 3

[0042]

[0043] The dynamic weighting factor calculation process is as follows:

[0044] Table 4

[0045] As shown in Table 4, the visual judgment of the target type is inconsistent with the radar point cloud simulation. The radar-measured radial velocity of the target and the motion trend calculated by vision are consistent with the radar data in both direction and magnitude. The radar point cloud simulation of the obstacle path is consistent with the visual judgment of the obstacle path and lane curvature. Moreover, there is a conflict between the conclusions of any two sensors. Therefore, based on Table 2 above, the reward / penalty factor bj is calculated as follows: ; Based on the dynamic weighting factor and reward / penalty factor calculated above, the comprehensive confidence score is calculated as follows: ; Due to the calculated overall confidence level It is in the fourth zone, so the system performs active obstacle avoidance.

[0046] In one embodiment of the present invention, a high-speed sweeper obstacle avoidance system is also provided, including: a controller; The controller is configured to include: The information acquisition module is configured to acquire the detection results of each sensor when multiple sensors detect the same obstacle simultaneously. The calculation module is configured to dynamically adjust the confidence weights of each sensor based on real-time environmental information and vehicle operating mode, and determine the reward and penalty factors based on the detection results of each sensor, so as to calculate the comprehensive confidence based on the confidence weights and reward and penalty factors. The judgment module is configured to store threshold intervals and send corresponding decision instructions to the high-speed sweeper based on the threshold interval in which the comprehensive confidence level falls.

[0047] In one embodiment of the present invention, a computer program product is also provided, including a computer program / instructions that, when executed by a processor, implement the steps of the high-speed sweeper obstacle avoidance method.

[0048] In one embodiment of the present invention, a non-transitory computer-readable storage medium is also provided, storing computer device-readable instructions that, when executed by at least one processor, cause the steps of the high-speed sweeper obstacle avoidance method to be performed.

[0049] In this embodiment, the processor is equivalent to the controller described above. In some embodiments, the controller may be, but is not limited to, a PLC, an embedded control system, an industrial computer, a processor, etc.

[0050] Embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0051] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0052] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0053] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0054] The above description is only a preferred embodiment of the present invention. Without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for obstacle avoidance by a high-speed road sweeper, characterized in that, include: When the controller detects that multiple sensors simultaneously detect the same obstacle, it acquires the detection results from each sensor. The controller dynamically adjusts the confidence weights of each sensor based on real-time environmental information and vehicle operating mode, and determines the reward and penalty factors based on the detection results of each sensor, so as to calculate the comprehensive confidence level based on the confidence weights and reward and penalty factors. The controller is configured to store threshold intervals and send corresponding decision instructions to the high-speed sweeper based on the threshold interval in which the comprehensive confidence level falls.

2. The obstacle avoidance method for high-speed road sweepers according to claim 1, characterized in that, The real-time environmental information includes one or more of the following: weather conditions, lighting conditions, and vehicle speed.

3. The obstacle avoidance method for high-speed road sweepers according to claim 1 or 2, characterized in that, When multiple sensors detect the same obstacle simultaneously via the controller, before acquiring the detection results from each sensor, the following steps are also included: In response to a work instruction, the controller acquires vehicle operating status data and environmental perception data from multiple sensors; The controller identifies curves based on operating status data and environmental perception data; The controller determines the operating mode of the high-speed sweeper based on the operating status data and curve recognition results.

4. The obstacle avoidance method for high-speed road sweepers according to claim 3, characterized in that, The environmental perception data includes point cloud data, image data, and distance data; the vehicle operating status data includes steering wheel angle and vehicle speed. The controller performs curve recognition based on operating status data and environmental perception data, including: The road curvature is determined based on the point cloud data or the image data, and the road boundary is fitted to the point cloud data. When the road curvature is greater than a first preset threshold, the steering wheel angle is greater than a second preset threshold within a preset duration, and the road boundary fitting result is a curve, it is determined to be a curve.

5. The obstacle avoidance method for high-speed road sweepers according to claim 4, characterized in that, The controller determines the operating mode of the high-speed sweeper truck based on the operating status data and curve recognition results, including: Based on the current vehicle speed and curve recognition results, the operating mode is determined to be one of the following three: If the current vehicle speed is lower than the third preset threshold, it is determined to be a low-speed edge-keeping operation mode; If the current vehicle speed is higher than the third preset threshold and the curve recognition result is a curve, then it is determined to be a high-speed curve operation mode. If the current vehicle speed is higher than the third preset threshold and the curve recognition result is not a curve, then it is determined to be a straight road high-speed operation mode.

6. The obstacle avoidance method for high-speed road sweepers according to claim 5, characterized in that, In the high-speed curve operation mode, the dynamic adjustment of the confidence weights of each sensor includes: Reduce the corresponding weights of sensors inside the vehicle.

7. The obstacle avoidance method for high-speed road sweepers according to claim 1, characterized in that, The formula for calculating the overall confidence level is as follows: ； In the formula, To represent the overall confidence level, n is the number of sensors that detected the same target; ci is the independent confidence level of the i-th sensor for that target; wi is the dynamic weighting factor of the i-th sensor in the current environment; m is the number of applicable reward / penalty factors; bj is the j-th reward / penalty factor. ； In the formula, The base weights for the i-th sensor are: This is a weather adjustment factor. This is the illumination adjustment factor. This is the vehicle speed adjustment coefficient. Adjustment coefficient for vehicle operating mode; The determination of reward and penalty factors based on the detection results of each sensor includes: A reward value is assigned when the detection results of all sensors are consistent, and a penalty value is assigned when the detection results of all sensors conflict. The formula for calculating the reward and punishment factor is as follows: ； In the formula, Rewards are matched to the type. For consistent speed, Rewards are given for consistency in trajectory continuity. Penalty for sensor collisions.

8. A high-speed road sweeper, characterized in that, include: Vehicle body; The sensing unit includes multiple sensors for acquiring point cloud data, image data, and distance data of the front and sides of the vehicle body; A controller, which is signal-connected to a sensing unit, is configured to perform the obstacle avoidance method for a high-speed sweeper as described in any one of claims 1-7.

9. A high-speed road sweeper obstacle avoidance system, characterized in that, include: Controller; The controller is configured to include: The information acquisition module is configured to acquire the detection results of each sensor when multiple sensors detect the same obstacle simultaneously. The calculation module is configured to dynamically adjust the confidence weights of each sensor based on real-time environmental information and vehicle operating mode, and determine the reward and penalty factors based on the detection results of each sensor, so as to calculate the comprehensive confidence level based on the confidence weights and reward and penalty factors. The judgment module is configured to store threshold intervals and send corresponding decision instructions to the high-speed sweeper based on the threshold interval in which the comprehensive confidence level falls.

10. A computer program product, characterized in that, Includes a computer program / instruction that, when executed by a processor, implements the steps of the obstacle avoidance method for a high-speed sweeper as described in any one of claims 1 to 7.

11. A non-transitory computer-readable storage medium, characterized in that, The device stores computer-readable instructions that, when executed by at least one processor, cause the steps of the obstacle avoidance method for a high-speed sweeper as described in any one of claims 1 to 7 to be performed.

Images