Sweeper path planning method and device based on multiple control centers

Abstract

The invention discloses a sweeping vehicle path planning method and device based on multiple control centers, relates to the technical field of sweeping vehicle path planning, and provides the following scheme that a dual-mode control system is configured for a cleaning vehicle, and the dual-mode control system comprises an independent driving control center and an independent sweeping control center; establishing a driving control coordinate system based on the center of a rear axle of the vehicle, establishing a sweeping control coordinate system according to a sweeping inertia compensation result, and generating an inertia compensation projection point as a transition coordinate for dual-control mode switching; the road environment is sensed in real time, and region types are dynamically divided according to the obstacle distance and the road width. A driving and sweeping double-coordinate-system independent control framework is adopted, a driving coordinate system is established based on the center of a rear axle of the vehicle, a sweeping coordinate system accurately reflects the operation posture by dynamically compensating sweeping inertia offset, switching between the driving coordinate system and the sweeping coordinate system is achieved through inertia compensation projection points, and the sweeping path precision is improved; the problem of path deviation caused by a traditional single coordinate system is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of cleaning vehicle path planning, and specifically relates to a path planning method and device for a cleaning vehicle based on multiple control centers. Background Art

[0002] Traditional cleaning vehicle path planning methods generally adopt a rigid body kinematic model based on a single fixed reference point, such as the vehicle's centroid or the center of the rear axle, to generate a global path through a unified coordinate system. This planning method based on a single fixed reference has high calculation efficiency and is relatively simple to implement, and performs well in standard configuration vehicles such as passenger cars, but it is difficult to adapt to the special operation requirements of cleaning vehicles.

[0003] Conventional path planning methods use a fixed reference point to establish a single coordinate system, which cannot take into account the dual requirements of driving stability and cleaning accuracy, and cannot accurately reflect the pose changes during the operation of the sweeping brush. Systemic path deviations occur under steering or rapid extension conditions. Therefore, we propose a path planning method and device for a cleaning vehicle based on multiple control centers to solve this problem. Summary of the Invention

[0004] To solve the above technical problems, a path planning method and device for a cleaning vehicle based on multiple control centers are provided. The technical solution of the present invention solves the problems in the above background art that the rigid body kinematic modeling using a single coordinate reference cannot adapt to the non-rigid characteristics of the dynamic extension of the sweeping mechanism of the cleaning vehicle, resulting in a deviation between the planned path and the actual cleaning requirements; the planning system based on fixed control points is difficult to track the change of the sweeping brush position in real time, affecting the path tracking accuracy; and the use of a unified collision detection weight lacks the ability to adaptively adjust according to environmental characteristics and the state of the sweeping brush.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: The present invention provides a path planning method for a cleaning vehicle based on multiple control centers, and the method includes: Step 1: Configure a dual-mode control system for the cleaning vehicle, including an independent driving control center and a cleaning control center; Step 2: Establish a driving control coordinate system based on the center of the rear axle of the vehicle, and at the same time establish a cleaning control coordinate system according to the sweeping brush inertia compensation result, and generate an inertia compensation projection point as the transition coordinate for the switching of the dual control mode; Step 3: Sense the road environment in real time, dynamically divide the area type according to the obstacle distance and the road width, and differentially configure the collision weight and the inflation coefficient; Step 4: Based on the collaborative work of the dual control centers, comprehensively apply the configured collision weight and inflation coefficient to generate a feasible operation path for the cleaning vehicle in stages.

[0006] Furthermore, a path planning device for a sweeper based on multiple control centers is proposed, which is used to implement the path planning method for a sweeper based on multiple control centers described in any one of the above, including: A dual-mode control module, which is used to configure a dual-mode control system for the cleaning vehicle, including an independent driving control center and a cleaning control center; A multi-source information processing module, which is used to establish a driving control coordinate system based on the center of the rear axle of the vehicle, and at the same time establish a cleaning control coordinate system according to the sweeping brush inertia compensation result, and generate an inertia compensation projection point as the transition coordinate for the switching of the dual control modes; An environment adaptive parameter configuration module, which is used to sense the road environment in real time, dynamically divide the area type according to the obstacle distance and the road width, and differentially configure the collision weight and the expansion coefficient; A multi-modal path generation module, which is used to generate a feasible operation path for the cleaning vehicle in stages based on the collaborative work of the dual control centers, and comprehensively apply the configured collision weight and expansion coefficient.

[0007] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention adopts an independent control architecture of a driving and a cleaning dual coordinate system. The driving coordinate system is established based on the center of the rear axle of the vehicle to ensure driving stability. The cleaning coordinate system accurately reflects the operation pose by dynamically compensating the inertial offset of the sweeping brush. The two are switched through the inertial compensation projection point, which improves the accuracy of the cleaning path and effectively solves the path deviation problem caused by the traditional single coordinate system. By sensing the road environment in real time, dynamically adjusting the collision weight and the expansion coefficient, and combining the garbage coverage rate to trigger the mode switching, the optimal path planning for the whole scene is realized, the success rate of edge cleaning is improved, and it automatically returns to the original position after avoiding obstacles, significantly improving the operation efficiency and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 is a flowchart of a path planning method for a sweeper based on multiple control centers proposed by the present invention; Figure 2 is a flowchart of the method for dynamically configuring control parameters in the present invention; Figure 3 is a structural block diagram of a path planning device for a sweeper based on multiple control centers proposed by the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0009] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations.

[0010] Refer to Figures 1-3 as shown, a path planning method for a sweeper based on multiple control centers includes: Step 1: Configure a dual-mode control system for the cleaning vehicle, including an independent driving control center and a sweeping control center; Step 2: Establish a driving control coordinate system based on the center of the vehicle's rear axle. At the same time, establish a sweeping control coordinate system according to the sweeping brush inertia compensation result, and generate an inertia compensation projection point as the transition coordinate for the switching of the dual control modes; Step 3: Sense the road environment in real time, dynamically divide the area types according to the obstacle distance and the road width, and differentially configure the collision weight and the expansion coefficient; Step 4: Based on the collaborative work of the dual control centers, comprehensively apply the configured collision weight and expansion coefficient, and generate the feasible operation path of the cleaning vehicle in stages. In the embodiment of the present invention, the establishment of the driving control coordinate system based on the center of the vehicle's rear axle, and the establishment of the sweeping control coordinate system according to the sweeping brush inertia compensation result, and the generation of the inertia compensation projection point as the transition coordinate for the switching of the dual control modes specifically include: Collect the real-time pose information of the vehicle chassis parameters and the sweeping brush device; According to the wheelbase parameter, calculate the midpoint coordinates of the line connecting the center points of the two hubs on both sides, which is defined as the rear axle center reference point, and construct a right-handed coordinate system with this reference point as the origin as the driving control center coordinate system; Based on the sweeping brush pose information, extract the real-time geometric center points of each sweeping brush, calculate the inertia offset vector during the sweeping brush extension process through the kinematic model, and perform inertia compensation on the sweeping brush geometric center points; Adopt the arithmetic mean method for the compensated sweeping brush geometric center coordinates to solve the overall geometric center coordinates of multiple sweeping brushes, and establish a right-handed coordinate system with the overall geometric center coordinates as the sweeping control center coordinate system; According to the change amount of the sweeping brush pose, generate the inertia compensation projection point of the rear axle center, and use this projection point as the transition reference point for the switching from the driving control reference system to the sweeping control reference system.

[0011] In the embodiment of the present invention, the obtaining of the inertia offset vector during the sweeping brush extension process through kinematic calculation and the compensation of the sweeping brush geometric center point specifically include: Obtain the current position of each sweeping brush , the fully extended position of the sweeping brush and the maximum inertia offset angle ; Obtain the sweeping brush extension speed , the maximum extension speed and the sweeping brush direction vector, and calculate the normalized direction based on the sweeping brush direction vector ; Specifically, obtain the sweeping brush extension speed through the built-in encoder of the sweeping brush drive mechanism, obtain the fixed offset of the sweeping brush mounting base relative to the center of the rear axle, obtain the current angle through the joint encoder of the sweeping brush, calculate the coordinates of the end in the local coordinate system of the sweeping brush in combination with the length of the sweeping brush arm, and transform it to the driving control center coordinate system through the homogeneous transformation matrix to obtain the current position of the sweeping brush, and calculate the direction vector of the sweeping brush; control the sweeping brush to perform a full-stroke extension movement with the maximum driving force, use the built-in encoder to record the shortest completion time as the minimum time for full extension, and calculate the maximum extension speed in combination with the length of the sweeping brush arm. The joint encoder records the current angle, and obtains the fully extended position of the sweeping brush in the same way as obtaining the current position of the sweeping brush. At the same time, install an inertial measurement unit at the end of the sweeping brush, and obtain the maximum inertial offset angle by integrating the angular velocity data; Construct the convex hull of the current position and the fully extended position of the sweeping brush, and extract the convex hull vertex set , from the formula Calculate the leading edge point ; From the formula Calculate the inertial compensation vector of each sweeping brush, where, is the inertial compensation vector, is the geometric offset, is the speed weight; From the formula Calculate the geometric center point of the sweeping brush after inertial compensation, where, is the current geometric center point of the sweeping brush.

[0012] Refer to Figure 2 As shown, generating the inertial compensation projection point of the rear axle center based on the sweeping brush pose information specifically includes: Quantify the positional relationship between the two into a coupling factor by calculating the ratio of the distance between the inertial compensation center point and the rear axle center to the wheelbase; Obtain the current extension length of the sweeping brush, calculate the ratio of the current sweeping brush extension length to the maximum extension length of the sweeping brush, and use the arithmetic mean method to calculate the overall extension ratio , linearly map and adjust the coupling factor based on the overall extension ratio to generate a dynamic coupling factor ; It should be noted that analyze the dynamic coupling factor from the formula , where, is the maximum coupling factor when the sweeping brush is fully extended, is the basic coupling factor when the sweeping brush is fully retracted.

[0013] Obtain the vehicle's real-time steering angle and the center of the vehicle's rear axle r, combined with the dynamic coupling factor, from the formula Calculate the projection center point as the inertial compensation projection point, where, is the projection center point, is the wheelbase of the vehicle.

[0014] In the embodiment of the present invention, the control parameters are dynamically configured according to the real-time road environment. The control parameters include a collision weight and a dilation coefficient, and specifically include: Obtain the vehicle width , the distance to the nearest obstacle currently and the road width ; When it is detected that the distance to the nearest obstacle currently is greater than half of the road width, the current road environment is determined as a square area. When it is detected that the distance to the nearest obstacle currently is less than or equal to half of the road width, the current road environment is determined as a narrow road area. Based on the narrow road area, when it is detected that the distance to the obstacle is less than 0.3 meters and it is a fixed obstacle, the road environment on that side is determined as a side area; According to the calculation model Analyze the basic collision weight of the cleaning vehicle in different road environments, where is the basic collision weight, is the basic collision weight in the open area, is the safety standard collision weight; Referentially, according to the safety standard of industrial vehicles, the value of the basic collision weight in the open area is 0.2, and the value of the safety standard collision weight is 0.7; It should be noted that means that the obstacle is outside the road boundary, belonging to a completely open square area, and directly adopt the value of the basic collision weight, means that the obstacle is within the road boundary, and the weight needs to be dynamically adjusted according to the distance.

[0015] When it is determined as the side area, the collision weight on the side where the cleaning vehicle is close to the roadside is set to , and the collision weight on the other side is set to ; Set the vehicle size in the vehicle algorithm according to the road environment, and record the ratio of the vehicle size in the vehicle algorithm to the vehicle width as the dilation coefficient.

[0016] In the embodiment of the present invention, based on the control center, combining the collision weight and the dilation coefficient, generate a feasible operation path for the cleaning vehicle, specifically including: Enable the driving control center at the starting stage of the cleaning vehicle, establish an environment perception system based on multi-sensor fusion, collect road topology information in real time, and dynamically adjust the dilation coefficient based on the road topology information and vehicle parameters ; From the formula Calculate the dilation coefficient , where The maximum expansion coefficient is 1.5, and the minimum expansion coefficient is 1.1; Based on the driving control center coordinate system, with the road center line as the reference trajectory, path offset is carried out away from the roadside based on the minimum mandatory interval distance, and the offset path is smoothed and optimized to generate a driving path; from the formula Analyze the minimum mandatory interval distance between the cleaning vehicle and the roadside, where is the minimum mandatory interval distance, and is the safety margin constant; Obtain the garbage distribution data within a predetermined range in front of the cleaning vehicle, calculate the real-time garbage coverage rate. If it is detected that the garbage coverage rate of the road surface for a continuous preset length exceeds the set threshold, the cleaning vehicle switches to the cleaning mode, and the expansion coefficient returns to the initial value; Specifically, scan the area 10 meters in front of the vehicle through a lidar to obtain three-dimensional point cloud data, and segment the ground effective point cloud. At the same time, a vision sensor collects images of the same area, and uses an object detection algorithm to identify garbage targets in the images and output confidence levels; spatially align the effective point cloud with the vision detection results and map them into a preset 20×20 ground grid; for each grid cell, perform weighted calculation by combining the point cloud density and the confidence level to obtain the garbage density value of the grid cell; calculate the number of grid cells with a garbage density value greater than 0.4 g / cm² among all grid cells, and the ratio of this number to the total number of grids is the garbage coverage rate. When the garbage coverage rate is greater than 0.7 within 3 consecutive detection cycles, switch the vehicle operating condition to the cleaning mode; obtain the geometric feature parameters of the roadside through the environmental perception system, and generate a cleaning path parallel to the roadside in combination with the cleaning control center coordinate system; Based on the road environment and collision weights, combined with the road topology information, generate an obstacle avoidance path. When the cleaning vehicle passes the obstacle, it automatically switches back to the cleaning path; Specifically, based on the environmental perception system, perform three-dimensional modeling on the obstacle, and accurately calculate its safety envelope range; adjust the safety expansion radius based on the collision weight and vehicle speed, and establish a multi-dimensional constraint space including collision cost, curvature limit, and dynamic constraints; use the RRT path search algorithm to solve the optimal obstacle avoidance path within the constraint space.

[0017] Refer to Figure 3 As shown, this solution proposes a cleaning vehicle path planning device based on multiple control centers for implementing the above-mentioned cleaning vehicle path planning method based on multiple control centers, including: A dual-mode control module for configuring a dual-mode control system for the cleaning vehicle, including an independent driving control center and a cleaning control center; The multi-source information processing module is used to establish a driving control coordinate system based on the center of the vehicle's rear axle, and at the same time establish a cleaning control coordinate system according to the sweeping brush inertia compensation result, and generate an inertia compensation projection point as the transition coordinate for the switching of the dual control modes; The environmental adaptive parameter configuration module is used to sense the road environment in real time, dynamically divide the area types according to the obstacle distance and the road width, and differentially configure the collision weight and the inflation coefficient; The multi-modal path generation module is used to generate a feasible operation path for the cleaning vehicle in stages based on the collaborative work of the dual control centers, and comprehensively apply the configured collision weight and inflation coefficient.

[0018] The multi-source information processing module specifically includes: The vehicle chassis information acquisition unit is used to obtain the wheel speed, steering angle and IMU data of the cleaning vehicle; The sweeping brush pose detection unit is used to collect the joint angles and extension states of each sweeping brush.

[0019] The environmental adaptive parameter configuration module specifically includes: The road environment feature recognition unit is used to monitor the ratio relationship between the obstacle distance and the road width in real time, and divide the road environment into three categories: square area, narrow road area and edge area; The collision weight calculation unit is used to set different collision weights for different road environments.

[0020] The multi-modal path generation module specifically includes: The driving path generation unit is used to calculate the safety offset by the minimum forced interval distance, and take the road center line as the benchmark, and optimize the trajectory by cubic spline interpolation to generate the driving path; The cleaning path decision unit is used to detect the pavement garbage distribution density in real time. When the coverage rate of consecutive sections exceeds the set threshold, it triggers the mode switching, resets the inflation coefficient to the initial value, and at the same time generates an operation path parallel to the road edge; The obstacle avoidance path generation unit is used to dynamically evaluate the obstacle collision risk, generate a temporary obstacle avoidance path in the constraint space, and automatically resume the original operation path after passing the obstacle.

[0021] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A path planning method for a road sweeper based on multiple control centers, characterized in that: The method comprises: Step 1: Configure a dual-mode control system for the cleaning vehicle, including an independent driving control center and a cleaning control center; Step 2: Establish a driving control coordinate system based on the center of the vehicle's rear axle, and establish a sweeping control coordinate system based on the sweeping brush inertia compensation result, and generate an inertia compensation projection point as a transition coordinate for switching between dual control modes; Step 3: Real-time perception of the road environment, dynamic division of area types according to obstacle distance and road width, and differential configuration of collision weights and expansion coefficients; Step 4: Based on the collaborative work of the dual control centers, the configured collision weights and expansion coefficients are comprehensively applied to generate feasible operation paths for the cleaning vehicle in stages.

2. The method according to claim 1, characterized in that A driving control coordinate system is established based on the center of the vehicle's rear axle. At the same time, a sweeping control coordinate system is established based on the sweeping brush inertia compensation result, and an inertia compensation projection point is generated as the transition coordinate for switching between dual control modes. Specifically, it includes: Collect vehicle chassis parameters and real-time position information of the sweeping device; According to the wheelbase parameters, the coordinates of the midpoint of the line connecting the center points of the wheel hubs on both sides are calculated and defined as the rear axle center reference point. A right-handed coordinate system is constructed with this reference point as the origin as the driving control center coordinate system. Based on the brush posture information, the real-time geometric center point of each brush is extracted, the inertial offset vector during the brush extension process is calculated through the kinematic model, and the inertia compensation of the brush geometric center point is performed; The geometric center coordinates of the brushes after compensation are calculated by arithmetic mean method to solve the overall geometric center coordinates of the multiple brushes, and a right-handed coordinate system is established with the overall geometric center coordinates as the cleaning control center coordinate system; According to the change of sweeping brush posture, the inertia compensation projection point of the rear axle center is generated, and the projection point is used as the transition reference point for switching from the driving control reference system to the sweeping control reference system.

3. The method according to claim 2, characterized in that The inertial offset vector of the sweep brush during the extension process is obtained by kinematic solution, and the geometric center point of the sweep brush is compensated, specifically including: Get the current position of each brush , brush fully extended position and the maximum inertial deviation angle ; Get the brush extension speed , Maximum extension speed and the sweep direction vector, and calculate the normalized direction based on the sweep direction vector ; Construct the convex hull of the current position and the fully extended position of the brush, and extract the convex hull vertex set , according to the formula Calculate frontier points ; By formula Calculate the inertia compensation vector of each brush, where: is the inertia compensation vector, is the geometric offset, is the speed weight; By formula Analyze the brush geometric center point after inertia compensation, where: The current geometric center point of the brush.

4. The method according to claim 3, characterized in that The step of generating the inertia compensation projection point of the rear axle center based on the sweeping posture information specifically includes: By calculating the ratio of the distance between the inertia compensation center point and the rear axle center to the wheelbase, the positional relationship between the two is quantified as a coupling factor; Get the current extension length of the brush, calculate the ratio of the current extension length to the maximum extension length of the brush, and use the weighted average method to calculate the overall extension ratio , adjust the coupling factor based on the linear mapping of the overall stretch ratio to generate a dynamic coupling factor ; Get the vehicle's real-time steering angle and the center of the vehicle's rear axle r, combined with the dynamic coupling factor, given by the formula Calculate the projection center point as the inertia compensation projection point, where, is the projection center point, is the vehicle wheelbase.

5. The method according to claim 1, characterized in that The control parameters are dynamically configured according to the real-time road environment, and the control parameters include collision weight and expansion coefficient, and specifically include: Get vehicle width , Current distance to the nearest obstacle and road width ; When the distance to the nearest obstacle is greater than half the width of the road, the current road environment is determined as a square area. When the distance to the nearest obstacle is less than or equal to half the width of the road, the current road environment is determined as a narrow road area. Based on the narrow road area, when the distance to the obstacle is less than 0.3 meters and it is a fixed obstacle, the road environment on this side is determined as an edge area. According to the calculation model Analyze the basic collision weight of the cleaning vehicle in different road environments, among which, is the base collision weight, is the basic collision weight of the open area, is the safety standard collision weight; When it is determined to be a curb area, the collision weight of the cleaning vehicle close to the curb is set to , the collision weight on the other side is set to ; The vehicle size in the vehicle algorithm is set according to the road environment, and the ratio of the vehicle size in the vehicle algorithm to the vehicle width is recorded as the expansion coefficient.

6. The method according to claim 5, characterized in that The generation of a feasible operation path of the cleaning vehicle based on the control center, combined with the collision weight and the expansion coefficient, specifically includes: In the initial stage of the cleaning vehicle, the driving control center is activated to establish an environmental perception system based on multi-sensor fusion, collect road topology information in real time, and dynamically adjust the expansion coefficient based on road topology information and vehicle parameters. ; Based on the driving control center coordinate system and taking the road centerline as the reference trajectory, the path is offset to the side away from the roadside based on the minimum mandatory interval distance, and the offset path is smoothed and optimized to generate the driving path; the formula is used Analyze the minimum mandatory separation distance between the cleaning vehicle and the curb, where: is the minimum mandatory separation distance, is the safety margin constant; Obtain the garbage distribution data within the predetermined range in front of the cleaning vehicle and calculate the real-time garbage coverage rate. If it is detected that the garbage coverage rate of the road surface for a continuous preset length exceeds the set threshold, the cleaning vehicle switches to the cleaning mode and the expansion coefficient returns to the initial value; The geometric characteristic parameters of the curb are obtained through the environmental perception system, and combined with the coordinate system of the cleaning control center, a cleaning path parallel to the curb is generated; Based on the road environment and collision weight, combined with the road topology information, an obstacle avoidance path is generated. When the cleaning vehicle passes the obstacle, it automatically switches back to the cleaning path.

7. A path planning device for a sweeper based on multiple control centers, characterized in that: The method for implementing a path planning method for a road sweeper based on multiple control centers as described in any one of claims 1 to 6 comprises: A dual-mode control module is used to configure a dual-mode control system for the cleaning vehicle, including an independent driving control center and a cleaning control center; A multi-source information processing module is used to establish a driving control coordinate system based on the center of the rear axle of the vehicle, and to establish a sweeping control coordinate system according to the sweeping brush inertia compensation result, and to generate an inertia compensation projection point as a transition coordinate for switching between dual control modes; Environmental adaptive parameter configuration module, which is used to perceive the road environment in real time, dynamically divide the area type according to the obstacle distance and road width, and configure the collision weight and expansion coefficient differently; The multimodal path generation module is used to generate feasible operation paths for the cleaning vehicle in stages based on the collaborative work of the dual control centers and the comprehensive application of configured collision weights and expansion coefficients.

8. The path planning device for a road sweeper based on multiple control centers according to claim 7, characterized in that: The multi-source information processing module specifically includes: Vehicle chassis information collection unit, used to obtain clean wheel speed, steering angle and IMU data; The sweeping brush posture detection unit is used to collect the angle and extension status of each sweeping brush joint.

9. The path planning device for a road sweeper based on multiple control centers according to claim 7, characterized in that: The environment adaptation parameter configuration module specifically includes: The ring road feature recognition unit is used to monitor the ratio of obstacle distance to road width in real time, and divide the road environment into three categories: square area, narrow road area and edge area; The collision weight calculation unit is used to set differentiated collision weights for different road environments.

10. The path planning device for a road sweeper based on multiple control centers according to claim 7, characterized in that: The multimodal path generation module specifically includes: A driving path generation unit is used to calculate the safety offset through the minimum mandatory separation distance, and to generate a driving path by optimizing the trajectory using cubic spline interpolation based on the road centerline; The cleaning path decision unit is used to detect the density of road garbage distribution in real time. When the coverage rate of continuous road sections exceeds the set threshold, the mode switching is triggered, the expansion coefficient is reset to the initial value, and an operation path is generated that is parallel to the road edge; The obstacle avoidance path generation unit is used to dynamically evaluate the obstacle collision risk, generate a temporary obstacle avoidance path within the constrained space, and automatically restore to the original operating path after passing the obstacle.

Images