A robot dynamic obstacle avoidance method based on efficiency evaluation and strategic fallback

Abstract

The application discloses a kind of robot dynamic obstacle avoidance methods based on efficiency evaluation and strategic retreat, method is: in the process that robot moves along the path planned by current global path, real-time perception environment carries out local dynamic obstacle avoidance, while real-time calculation is carried out for the comprehensive efficiency index for quantifying path traffic efficiency;Comprehensive efficiency index is compared with preset efficiency threshold value, when comprehensive efficiency index is lower than efficiency threshold value, strategic retreat instruction is generated;According to strategic retreat instruction, robot is retreated to a safe local obstacle avoidance starting point from current position;In the process of retreat or after reaching local obstacle avoidance starting point, the area between current position and local obstacle avoidance starting point is marked as known inefficient area, and global path re-planning is executed, wherein the path of re-planning avoids known inefficient area.The application gives consideration to path efficiency and motion smoothness, which helps to reduce mechanical loss and energy consumption, and improve the stability and reliability of system operation.

Description

Technical Field

[0001] This invention relates to the field of robot obstacle avoidance technology, and in particular to a dynamic obstacle avoidance method for robots based on efficiency assessment and strategic retreat. Background Technology

[0002] With the widespread application of robots, the number of dynamic obstacles in their working environments is increasing, placing higher demands on real-time obstacle avoidance capabilities. While traditional obstacle avoidance solutions can ensure basic safety, in scenarios with dense dynamic obstacles, robots are prone to getting caught in a "local optimum trap," resulting in frequent adjustments and slow progress, leading to low overall task efficiency and increased time and energy consumption. Current mainstream technologies mostly focus on real-time adjustments to local paths, lacking a macro-level assessment of the efficiency of the entire path, making it difficult to make strategic adjustments when efficiency drops significantly.

[0003] Current mainstream technical solutions primarily employ a combination of global planning and local adjustments. Global path planning utilizes algorithms such as A*, while local obstacle avoidance is achieved through methods like Dynamic Windowing (DWA) and artificial potential fields. The core idea is to optimize local paths as closely as possible to the pre-defined global path while ensuring collision avoidance. However, most existing methods focus on "how to bypass the current obstacle" without strategically assessing "whether the entire path is still efficient." Existing methods lack effective real-time metrics to quantify the overall path's efficiency, failing to provide a reliable basis for strategic replanning. In congested environments with continuously appearing dynamic obstacles, local adjustments may lead to repeated adjustments within a small area, failing to fundamentally eliminate inefficient paths. Summary of the Invention

[0004] The purpose of this invention is to provide a robot dynamic obstacle avoidance method based on efficiency assessment and strategic retreat. By introducing a global efficiency assessment and strategic retreat mechanism, the robot's task completion efficiency on continuously congested paths is significantly improved.

[0005] The technical solution adopted in this invention is:

[0006] A robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat includes the following steps:

[0007] Path passage efficiency assessment: During the robot's movement along the path planned by the current global path, the robot perceives the environment in real time and performs local dynamic obstacle avoidance, while calculating the comprehensive efficiency index (EI) in real time to quantify the path passage efficiency.

[0008] Strategic rollback decision: The overall efficiency index (EI) is compared with a preset efficiency threshold. When the overall efficiency index falls below the efficiency threshold, a strategic rollback instruction is generated. The efficiency threshold is... It can be set to 0.6. When EI < At that time, a strategic retreat was triggered.

[0009] The evaluation period T, like the threshold, can be set manually or dynamically adjusted based on the complexity of the environment.

[0010] Retreat execution: Control the robot to retreat from its current position (Pcurrent) to a safe local obstacle avoidance starting point (Pstart) according to the strategic retreat command;

[0011] Path replanning: During the rollback process or after reaching the local obstacle avoidance starting point (Pstart), the area between the current position (Pcurrent) and the local obstacle avoidance starting point (Pstart) is marked as a known inefficient area, and global path replanning is performed, in which the replanned path avoids the known inefficient area.

[0012] Furthermore, the formula for calculating the Overall Efficiency Index (EI) is as follows:

[0013] ;

[0014] Where NVE stands for normalized speed efficiency. , To assess the actual average speed over the period, The ideal speed for obstacle-free operation within the assessment period; TAR represents the target approach rate. , and These represent the distances from the robot to the target point at the beginning and end of the evaluation cycle, respectively; S is the total actual path length of the robot during the evaluation cycle; and OER is the obstacle encounter rate. N represents the number of obstacles encountered during the evaluation period; MS represents motion stability. Ω represents the total rotation angle during the evaluation period, and A represents the total acceleration change during the evaluation period. , , and These are the weighting coefficients for standardized speed efficiency, target proximity rate, obstacle encounter rate, and motion stability, respectively. .

[0015] Specifically, the Overall Efficiency Index (EI) includes Standardized Velocity Efficiency (NVE), Target Approach Rate (TAR), Obstacle Encounter Rate (OER), and Motion Stability (MS). Standardized Velocity Efficiency (NVE) is the ratio of actual speed to ideal speed, reflecting whether movement is delayed. Target Approach Rate (TAR) assesses the effectiveness of progress towards the target point, avoiding unnecessary detours. Obstacle Encounter Rate (OER) reflects the smoothness of the path; the more obstacles encountered, the lower the score. Motion Stability (MS) assesses the smoothness of movement; frequent turns and accelerations / decelerations lower the score.

[0016] Furthermore, the efficiency threshold is either a fixed value preset according to task requirements or a variable dynamically adjusted according to environmental complexity. Specifically, as a feasible implementation, when the obstacle density detected by the robot's sensors exceeds a certain set threshold, the efficiency threshold is automatically lowered to make the robot more "patient"; conversely, the threshold is raised to make it retreat more decisively.

[0017] Furthermore, in the retreat execution step, the path for the robot to retreat to the local obstacle avoidance starting point is one of the following: retreat along a straight line, retreat along historical movement trajectory points, or plan a safe retreat path based on real-time environmental information.

[0018] Furthermore, the inefficient region is a polygonal region containing the current position (Pcurrent) and the local obstacle avoidance starting point (Pstart); specifically, as a feasible implementation, the inefficient region is a rectangular region, or a strip-shaped region with a certain width centered on the path, or a region generated based on obstacle clustering. The length and width of the rectangle are set as variables, which can be set or dynamically adjusted.

[0019] This invention employs the above technical solution to elevate the decision-making level of robot path planning from simple real-time obstacle avoidance to a macro-strategic assessment of the overall path's efficiency. When efficiency falls below a threshold, a strategic retreat is triggered, enabling the robot to proactively escape local optima and fundamentally solve the problem of repeated adjustments and low efficiency in dense, dynamic obstacle environments. By constructing a comprehensive efficiency index (EI) model that integrates multiple dimensions such as speed efficiency, target proximity rate, obstacle encounter rate, and motion stability, objective and quantitative data support is provided for retreat decisions, overcoming the shortcomings of traditional methods that rely on subjective or single-indicator judgments. After retreating, the robot proactively marks and avoids identified inefficient areas, thereby planning a completely new, efficient path, rather than patching up existing inefficient paths locally, significantly improving overall task efficiency in continuously congested environments.

[0020] This invention balances path efficiency and motion smoothness, which helps reduce mechanical losses and energy consumption, and improves the stability and reliability of system operation. Attached Figure Description

[0021] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments;

[0022] Figure 1 This is a flowchart illustrating a robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat according to the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0024] like Figure 1 As shown, this invention discloses a dynamic obstacle avoidance method for robots based on efficiency evaluation and strategic retreat, comprising the following steps:

[0025] Path passage efficiency assessment: During the robot's movement along the path planned by the current global path, the robot perceives the environment in real time and performs local dynamic obstacle avoidance, while calculating the comprehensive efficiency index (EI) in real time to quantify the path passage efficiency.

[0026] Strategic rollback decision: The overall efficiency index (EI) is compared with a preset efficiency threshold. When the overall efficiency index falls below the efficiency threshold, a strategic rollback instruction is generated. The efficiency threshold is... It can be set to 0.6. When EI < At that time, a strategic retreat was triggered.

[0027] The evaluation period T, like the threshold, can be set manually or dynamically adjusted based on the complexity of the environment.

[0028] Retreat execution: Control the robot to retreat from its current position (Pcurrent) to a safe local obstacle avoidance starting point (Pstart) according to the strategic retreat command;

[0029] Path replanning: During the backtracking process or after reaching the local obstacle avoidance starting point (Pstart), the area between the current position (Pcurrent) and the local obstacle avoidance starting point (Pstart) is marked as a known inefficient area, and global path replanning is performed to reach the destination. The replanned path avoids the known inefficient area. The above method uses precise backtracking to the position where obstacle avoidance was last started, rather than the starting point of the route, which effectively saves time.

[0030] Furthermore, the formula for calculating the Overall Efficiency Index (EI) is as follows:

[0031] ;

[0032] in, , , and These are the weighting coefficients for standardized speed efficiency, target proximity rate, obstacle encounter rate, and motion stability, respectively. .

[0033] NVE stands for Standardized Speed ​​Efficiency. , To assess the actual average speed over the period, To evaluate the ideal speed for obstacle-free operation within the cycle; the closer the NVE is to 1, the closer the robot's movement speed is to the ideal state, and the less frequent the delays.

[0034] TAR stands for target proximity rate. , and These represent the distances from the robot to the target point at the beginning and end of the evaluation cycle, respectively, and S is the total actual path length of the robot during the evaluation cycle; ideally, the robot moves towards the target in a straight line. Therefore, TAR≈1. Ineffective movements are caused by the robot frequently going around in circles or even moving in the opposite direction, resulting in... If so, then TAR=0.

[0035] OER stands for Obstacle Encounter Rate. N represents the number of obstacles encountered during the evaluation period; in the ideal case, N=0 and OER=1. The more obstacles encountered, the lower the OER. OER reflects the smoothness of the path.

[0036] MS stands for motion stability. Ω represents the total rotation angle within the evaluation period. , |ω| represents the absolute value of the angular velocity, and the duration of this angular velocity is Δt; A is the total change in acceleration during the evaluation period. , The absolute value of acceleration, and the duration of the acceleration Δt; if it is uniform linear motion, MS=1, and if there are frequent turns or accelerations and decelerations, MS tends to be 0.

[0037] Specifically, the Overall Efficiency Index (EI) includes Standardized Velocity Efficiency (NVE), Target Approach Rate (TAR), Obstacle Encounter Rate (OER), and Motion Stability (MS). Standardized Velocity Efficiency (NVE) is the ratio of actual speed to ideal speed, reflecting whether movement is delayed. Target Approach Rate (TAR) assesses the effectiveness of progress towards the target point, avoiding unnecessary detours. Obstacle Encounter Rate (OER) reflects the smoothness of the path; the more obstacles encountered, the lower the score. Motion Stability (MS) assesses the smoothness of movement; frequent turns and accelerations / decelerations lower the score.

[0038] Furthermore, the efficiency threshold is either a fixed value preset according to task requirements or a variable dynamically adjusted according to environmental complexity. Specifically, as a feasible implementation, when the obstacle density detected by the robot's sensors exceeds a certain set threshold, the efficiency threshold is automatically lowered to make the robot more "patient"; conversely, the threshold is raised to make it retreat more decisively.

[0039] Furthermore, in the retreat execution step, the path for the robot to retreat to the local obstacle avoidance starting point is one of the following: retreat along a straight line, retreat along historical movement trajectory points, or plan a safe retreat path based on real-time environmental information.

[0040] Furthermore, the inefficient region is a polygonal region containing the current position (Pcurrent) and the local obstacle avoidance starting point (Pstart); specifically, as a feasible implementation, the inefficient region is a rectangular region, or a strip-shaped region with a certain width centered on the path, or a region generated based on obstacle clustering. The length and width of the rectangle are set as variables, which can be set or dynamically adjusted.

[0041] This invention employs the above technical solution to elevate the decision-making level of robot path planning from simple real-time obstacle avoidance to a macro-strategic assessment of the overall path's efficiency. When efficiency falls below a threshold, a strategic retreat is triggered, enabling the robot to proactively escape local optima and fundamentally solve the problem of repeated adjustments and low efficiency in dense, dynamic obstacle environments. By constructing a comprehensive efficiency index (EI) model that integrates multiple dimensions such as speed efficiency, target proximity rate, obstacle encounter rate, and motion stability, objective and quantitative data support is provided for retreat decisions, overcoming the shortcomings of traditional methods that rely on subjective or single-indicator judgments. After retreating, the robot proactively marks and avoids identified inefficient areas, thereby planning a completely new, efficient path, rather than patching up existing inefficient paths locally, significantly improving overall task efficiency in continuously congested environments.

[0042] This invention balances path efficiency and motion smoothness, which helps reduce mechanical losses and energy consumption, and improves the stability and reliability of system operation.

[0043] Obviously, the described embodiments are only a part of the embodiments of this application, not all of them. Without conflict, the embodiments and features in the embodiments of this application can be combined with each other. The components of the embodiments of this application described and illustrated herein can generally be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of this application is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

Claims

1. A robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat, characterized in that: Includes the following steps: Path passage efficiency evaluation: During the robot's movement along the path planned by the current global path, the robot perceives the environment in real time and performs local dynamic obstacle avoidance, while simultaneously calculating a comprehensive efficiency index to quantify path passage efficiency. Strategic rollback decision: The overall efficiency index is compared with a preset efficiency threshold. When the overall efficiency index is lower than the efficiency threshold, a strategic rollback instruction is generated. The evaluation period T, like the threshold, can be set manually or dynamically adjusted based on the complexity of the environment. Retreat execution: Control the robot to retreat from its current position to a safe local obstacle avoidance starting point according to the strategic retreat command; Path replanning: During the backtracking process or after reaching the local obstacle avoidance starting point, the area between the current position and the local obstacle avoidance starting point is marked as a known inefficient area, and global path replanning is performed, in which the replanned path avoids the known inefficient area.

2. The robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat according to claim 1, characterized in that: The comprehensive efficiency index includes a weighted composite index of standardized speed efficiency (NVE), target proximity rate (TAR), obstacle encounter rate (OER), and motion stability (MS). in The formula for calculating the standardized velocity efficiency (NVE) is: , To assess the actual average speed over the period, To assess the ideal speed for unobstructed access within the evaluation period; The formula for calculating Target Approach Rate (TAR) is: , and These represent the distances from the robot to the target point at the beginning and end of the evaluation cycle, respectively, and S is the total length of the actual path moved by the robot during the evaluation cycle. The formula for calculating Obstacle Encounter Rate (OER) is: N represents the number of obstacles encountered during the evaluation period; The formula for calculating motion stability MS is: Ω represents the total rotation angle during the evaluation period, and A represents the total acceleration change during the evaluation period.

3. The robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat according to claim 2, characterized in that: The formula for calculating the overall efficiency index is: ； EI stands for Overall Efficiency Index. , , and These are the weighting coefficients for standardized speed efficiency, target proximity rate, obstacle encounter rate, and motion stability, respectively. .

4. The robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat according to claim 1, characterized in that: The efficiency threshold is either a fixed value preset according to task requirements, or a variable that is dynamically adjusted according to environmental complexity.

5. The robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat according to claim 4, characterized in that: When the robot's sensors detect an obstacle density exceeding a set threshold, the efficiency threshold is lowered; conversely, the threshold is raised.

6. The robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat according to claim 1, characterized in that: During the retreat execution step, the robot retreats to the local obstacle avoidance starting point in one of the following ways: retreating along a straight line, retreating along historical movement trajectory points, or planning a safe retreat path based on real-time environmental information.

7. The robot dynamic obstacle avoidance method based on efficiency evaluation and strategic retreat according to claim 1, characterized in that: The inefficient region is a polygonal region that includes the current position and the local obstacle avoidance starting point.

Images