A dynamic obstacle perception and emergency stop electronic control method for safe robot patrolling

Abstract

This invention belongs to the field of dynamic obstacle avoidance technology for robots, and provides a dynamic obstacle perception and emergency stop electronic control method for safe robot patrolling. This invention analyzes the emergency stop timing offset index between the dynamic obstacle and the robot using position data during their movement, reflecting the robot's emergency stop execution risk in the time dimension; and analyzes the robot's emergency stop execution consistency deviation coefficient, reflecting the spatial or amplitude consistency of the robot's emergency stop execution action; based on the emergency stop timing offset index and execution consistency deviation coefficient, it obtains the robot's emergency stop response reliability index, and then determines the robot's emergency stop trigger adjustment coefficient at the current moment, compensating for the target execution gain, and realizing the adjustment of the intensity and speed of the robot's drive execution unit's emergency stop response, ensuring that the robot's actions are fast, safe, and stable during the dynamic obstacle perception and emergency stop process.

Description

Technical Field

[0001] This invention relates to the field of dynamic obstacle avoidance technology for robots, and in particular to a dynamic obstacle perception and emergency stop control method for safe robot patrolling. Background Technology

[0002] With the increasing prevalence of autonomous mobile robots in warehousing and logistics, security patrols, and other scenarios, the environments they face are becoming increasingly complex. They often need to deal with dynamic obstacles such as pedestrians and vehicles that suddenly appear, placing extremely high demands on their active safety capabilities. Currently, most mainstream robot obstacle avoidance safety solutions are based on a central processing unit (CPU) software control architecture. Their typical working principle is as follows: sensor data is collected by the main controller, where software algorithms complete environmental perception, path planning, and generate motion control commands to achieve obstacle avoidance. Figure 2 and Figure 3 As shown.

[0003] However, existing electronic control systems typically rely on the upper controller to complete dynamic obstacle perception and risk assessment before issuing emergency stop commands to the drive execution unit via the software control link. This process is limited by factors such as sensor data refresh cycle, control task scheduling delay, and bus communication timing jitter, resulting in uncontrollable errors in the actual effective time of the emergency stop command under different operating conditions. When the dynamic obstacle approaches at a high speed or the robot's own motion state changes, timing deviations can directly cause safety hazards such as braking lag, inconsistent braking distance, or even emergency stop failure, thereby reducing the overall safety and reliability of the robot during patrol. Summary of the Invention

[0004] To address the above technical problems, this invention provides a dynamic obstacle perception and emergency stop electrical control method for safe robot patrolling.

[0005] According to the present invention, a dynamic obstacle perception and emergency stop electronic control method for safe robot patrolling is provided, the method comprising:

[0006] The position data of the dynamic obstacles and the robot at each moment during the movement are collected separately;

[0007] Based on the location data, the dynamic path convergence factor between the dynamic obstacle and the robot is analyzed, and combined with the emergency stop trigger delay time, the emergency stop timing offset index between the dynamic obstacle and the robot is obtained.

[0008] Based on the location data, the fluctuation of the robot's running acceleration during the process from emergency stop triggering to emergency stop completion is analyzed, and the consistency of the execution time from multiple historical emergency stop triggering to emergency stop completion is analyzed to obtain the robot's emergency stop execution consistency deviation coefficient.

[0009] The robot's emergency stop response reliability index is determined based on the emergency stop timing offset index and the emergency stop execution consistency deviation coefficient.

[0010] Based on the emergency stop response reliability index and the available distance between the robot and the dynamic obstacle at the current moment, the emergency stop trigger adjustment coefficient of the robot at the current moment is determined to compensate for the target execution gain.

[0011] In some embodiments of the present invention, analyzing the dynamic path convergence factor between dynamic obstacles and the robot based on the location data includes:

[0012] Preset time window;

[0013] Within a preset time window, based on the location data, the relative distance between the dynamic obstacle and the robot is obtained, as well as the movement direction of the dynamic obstacle and the robot respectively.

[0014] Within a preset time window, the dynamic approach time margin between the dynamic obstacle and the robot is analyzed based on the relative distance.

[0015] Within a preset time window, based on the direction of motion, the directional convergence trend between the dynamic obstacle and the robot is analyzed, and the stability of the dynamic obstacle's direction of motion is analyzed. Combined with the dynamic approach time margin, the dynamic path convergence factor between the dynamic obstacle and the robot within the preset time window is determined.

[0016] In some embodiments of the present invention, within a preset time window, the dynamic approach time margin between the dynamic obstacle and the robot is analyzed based on the relative distance, including:

[0017] Within a preset time window, the approach rate between the dynamic obstacle and the robot is obtained based on the relative distance;

[0018] Within a preset time window, based on the approach rate, the rate of change of the approach rate between the dynamic obstacle and the robot at adjacent sampling times is analyzed. Combined with the average approach rate within the preset time window, the dynamic approach time margin between the dynamic obstacle and the robot is obtained.

[0019] In some embodiments of the present invention, within a preset time window, based on the direction of motion, the convergence trend between the dynamic obstacle and the robot and the stability of the dynamic obstacle's motion direction are analyzed, including:

[0020] Within a preset time window, the angle between the dynamic obstacle and the robot is calculated based on the direction of motion. The changing trend of the angle of motion at each sampling moment is analyzed to obtain the directional convergence trend between the dynamic obstacle and the robot within the preset time window.

[0021] Within a preset time window, based on the motion direction of the dynamic obstacle, the change in the motion direction of the dynamic obstacle at adjacent sampling times is obtained, thus obtaining the stability of the motion direction of the dynamic obstacle.

[0022] In some embodiments of the present invention, the changing trend of the angle between the motion directions at each sampling moment is analyzed to obtain the directional convergence trend between the dynamic obstacle and the robot within a preset time window, including:

[0023] Calculate the ratio between the angle of motion direction at any sampling time and the angle of motion direction at the next sampling time to obtain the trend of change of the angle of motion direction at each sampling time;

[0024] By iterating through all sampling moments within a preset time window, the average value of the ratios between the angles of the motion directions is calculated to obtain the directional convergence trend between the dynamic obstacle and the robot within the preset time window.

[0025] In some embodiments of the present invention, the emergency stop timing offset index between the dynamic obstacle and the robot is obtained by combining the emergency stop trigger delay time, including:

[0026] The time difference between the emergency stop triggering time and the start of the emergency stop braking action is recorded as the emergency stop triggering delay time, and the average emergency stop triggering delay time corresponding to multiple historical emergency stop triggers is calculated.

[0027] Based on the dynamic path convergence factor and the average emergency stop trigger delay time, the emergency stop timing offset index between the dynamic obstacle and the robot is obtained.

[0028] In some embodiments of the present invention, based on the position data, the fluctuation of the robot's running acceleration during the process from triggering an emergency stop to completing the emergency stop is analyzed, including:

[0029] The time difference between the emergency stop trigger time and the emergency stop completion time is recorded as the emergency stop operation execution time.

[0030] Based on the location data, the robot's running acceleration at each sampling moment during the emergency stop operation execution time is obtained, and the standard deviation of all the running accelerations is calculated to obtain the fluctuation of the robot's running acceleration during the process from triggering the emergency stop to completing the emergency stop.

[0031] In some embodiments of the present invention, analyzing the consistency of execution time from multiple historical emergency stop triggers to the completion of the emergency stop includes:

[0032] Calculate the difference between the execution times of any two emergency stop triggers in history, iterate through all the difference values ​​corresponding to two emergency stop triggers in history, calculate the average value, and obtain the consistency in execution time from multiple emergency stop triggers to the completion of the emergency stop in history.

[0033] In some embodiments of the present invention, the emergency stop triggering adjustment coefficient of the robot at the current moment is determined based on the emergency stop response reliability index and the available distance between the robot and the dynamic obstacle at the current moment, including:

[0034] Obtain the minimum theoretical distance required from the current emergency stop trigger to the completion of the emergency stop;

[0035] Based on the location data, the available distance between the robot and the dynamic obstacle at the current moment is obtained;

[0036] Calculate the ratio between the available distance and the minimum theoretical distance to obtain the robot's emergency stop safety margin at the current moment;

[0037] Based on the emergency stop response reliability index and the emergency stop safety margin, the emergency stop trigger adjustment coefficient of the robot at the current moment is determined.

[0038] In some embodiments of the present invention, compensation for the target execution gain includes:

[0039] The maximum and minimum values ​​of the target execution gain parameters allowed by the electronic control system output to the emergency stop execution unit are obtained, and the target execution gain is compensated in combination with the emergency stop trigger adjustment coefficient.

[0040] As can be seen from the above embodiments, the dynamic obstacle perception and emergency stop control method for safe robot patrolling provided by the embodiments of the present invention has the following beneficial effects:

[0041] This invention analyzes the dynamic path convergence factor between dynamic obstacles and the robot by analyzing their position data during movement. Combined with the emergency stop trigger delay time, it obtains an emergency stop timing offset index between the dynamic obstacle and the robot, reflecting the emergency stop execution risk of the electronic control system in the time dimension. It also analyzes the fluctuation of the robot's acceleration during the emergency stop triggering and completion process, and the consistency of execution time across multiple historical emergency stop triggers and completions, obtaining an emergency stop execution consistency deviation coefficient, reflecting the spatial or amplitude consistency of the robot's emergency stop actions. Based on the emergency stop timing offset index and the emergency stop execution consistency deviation coefficient, it determines the robot's emergency stop response reliability index, reflecting the response timing delay and execution stability of the emergency stop command, providing a more unified risk assessment basis for dynamic obstacle perception emergency stop electronic control systems. Based on the emergency stop response reliability index and the available distance between the robot and the dynamic obstacle at the current moment, it determines the robot's emergency stop trigger adjustment coefficient at the current moment, compensating for the target execution gain, and adjusting the intensity and speed of the drive execution unit's emergency stop response, ensuring that the patrolling robot's actions are fast, safe, and stable during the dynamic obstacle perception emergency stop process.

[0042] This invention analyzes the temporal correlation between the approach trend of dynamic obstacles and the robot's own motion state, enabling the electronic control system to accurately identify the critical time interval where potential risks occur before emergency stop is triggered, thereby avoiding static judgment based solely on a single distance threshold; and by observing the response status of the emergency stop trigger command in the electronic control link, it can effectively identify potential lag or failure risks in the emergency stop execution process, thereby effectively improving the determinism of emergency stop triggering and the consistency of execution of the robot in complex dynamic environments.

[0043] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0044] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A schematic diagram of the basic process of a dynamic obstacle perception and emergency stop control method for safe robot patrolling provided in an embodiment of the present invention;

[0046] Figure 2 A schematic diagram of the overall framework of a robot control system provided by existing technology;

[0047] Figure 3 A schematic diagram illustrating a control implementation method for a robot control system provided by existing technology. Detailed Implementation

[0048] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a dynamic obstacle perception and emergency stop control method for safe robot patrolling proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device comprising a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of additional identical elements in the article or device that includes the element.

[0050] The scenario addressed by this invention is as follows: When an intelligent inspection robot in an industrial park is conducting patrol work, the robot continuously travels along a preset patrol path through areas such as factory passages, equipment areas, and intersections. Numerous dynamic obstacles with random motion characteristics exist in the surrounding area, including on-site workers, forklifts, material handling vehicles, and other mobile equipment. If a reliable emergency stop electrical control response cannot be triggered within a very short time, collisions or safety accidents can easily occur. Therefore, the purpose of this invention is to provide a dynamic obstacle perception and emergency stop electrical control method for safe robot patrolling, enabling the patrolling robot to move quickly, safely, and stably during the obstacle perception and emergency stop process.

[0051] The following will describe in detail, with reference to the accompanying drawings, a dynamic obstacle perception and emergency stop control method for safe robot patrolling provided in this embodiment.

[0052] Please see Figure 1 This illustrates the basic flow of a dynamic obstacle perception and emergency stop control method for safe robot patrolling provided by an embodiment of the present invention.

[0053] like Figure 1 As shown, an embodiment of the present invention provides a dynamic obstacle perception and emergency stop control method for safe robot patrolling, which specifically includes the following steps:

[0054] S100: Collects position data of dynamic obstacles and the robot at each moment during its movement.

[0055] During its patrol, the robot senses dynamic obstacles in front of and around the robot through its environmental perception unit. When a dynamic obstacle appears in the robot's field of vision, the robot's electronic control unit collects the position data and speed data of the dynamic obstacle and the robot at each moment during its movement. The position data and speed data are timestamped and used as input monitoring data for subsequent emergency stop risk assessment and electronic control response.

[0056] S200: Based on location data, analyze the dynamic path convergence factor between dynamic obstacles and the robot, and combine it with the emergency stop trigger delay time to obtain the emergency stop timing offset index between dynamic obstacles and the robot.

[0057] Based on location data, the dynamic path convergence factor between dynamic obstacles and the robot is analyzed, further including:

[0058] First, a preset time window is defined. Specifically, the preset time window covers multiple consecutive sampling moments. The preset time window can be shifted forward 10 consecutive sampling moments from the current sampling moment. If the dynamic obstacle has not yet entered the robot's field of vision when the current sampling moment is shifted forward 10 consecutive sampling moments, the preset time window can be shifted from the current sampling moment to the moment when the dynamic obstacle enters the robot's field of vision.

[0059] Then, within a preset time window, based on the position data, the relative distance between the dynamic obstacle and the robot is obtained, as well as the movement directions of the dynamic obstacle and the robot respectively. Specifically, within the preset time window, based on the position data of the dynamic obstacle and the robot at each moment during their movement, the relative distance (Euclidean distance) between the dynamic obstacle and the robot at each moment is calculated, thus obtaining a data set of the relative distance between the dynamic obstacle and the robot changing over time; and the movement direction of the dynamic obstacle (movement direction vector) is calculated by the change in the position coordinates of the dynamic obstacle at adjacent sampling moments, and the movement direction of the robot (movement direction vector) is calculated by the change in the position coordinates of the robot at adjacent sampling moments, thus obtaining data sequences of the movement directions of the dynamic obstacle and the robot changing over time respectively.

[0060] Furthermore, within a preset time window, the dynamic approach time margin between the dynamic obstacle and the robot is analyzed based on the relative distance. Specifically, within the preset time window, the approach rate between the dynamic obstacle and the robot at each sampling moment is obtained based on the relative distance. This is achieved by calculating the difference between the relative distances at each sampling moment and the previous sampling moment, and then comparing this difference with the time difference between the two sampling moments to obtain the approach rate between the dynamic obstacle and the robot at each sampling moment (if the approach rate is less than 0, indicating that the dynamic obstacle and the robot are far apart, the approach rate is set to 0). Within the preset time window, based on the approach rate, the rate of change of the approach rate between the dynamic obstacle and the robot at adjacent sampling moments is analyzed. This is achieved by calculating the ratio of the approach rate at each sampling moment to the approach rate at the next sampling moment to obtain the rate of change of the approach rate between the dynamic obstacle and the robot at each sampling moment. This process is repeated for all sampling moments within the preset time window to calculate the average rate of change of the approach rate. Combining the average rate of change of the approach rate and the average approach rate within the preset time window, the dynamic approach time margin between the dynamic obstacle and the robot is obtained. The formula for calculating the dynamic approach time margin between the dynamic obstacle and the robot is as follows:

[0061]

[0062] In the formula, This indicates the dynamic approach time margin between dynamic obstacles and the robot within a preset time window; This represents the average approach rate within a preset time window. Indicates the number of times within the preset time window The approach rate between the dynamic obstacle and the robot at each sampling time; Indicates the number of times within the preset time window The approach rate between the dynamic obstacle and the robot at each sampling time; This represents the total number of sampling moments within the preset time window; and Both represent denominator correction parameters, with the same dimensions as the denominator, and their values ​​are local minima greater than 0. and This is to prevent the denominator from being 0; for example, it can be set... , This can prevent the denominator from being 0 without significantly interfering with the calculation of normal values. This represents a linear normalization function, such as a max-min normalization function, used to normalize the dynamic approach time margin to... Within the range, the maximum value in the maximum-minimum normalization function is the preset upper limit of the range, which is set according to the maximum dynamic approach time that appears in the historical operating data.

[0063] Average approach rate The smaller the value, the lower the approach rate of the dynamic obstacle in the current operating state, and the greater the dynamic approach time margin between the dynamic obstacle and the robot. This represents the rate of change of the approach rate between the dynamic obstacle and the robot at adjacent sampling times. A larger value indicates that the approach rate between the dynamic obstacle and the robot decreases over time; dynamic approach time margin. The larger the value, the longer it takes for a dynamic obstacle to enter the robot's safe influence area under the current operating state, meaning a higher dynamic approach time margin.

[0064] Dynamic approach time margin reflects the timescale at which a dynamic obstacle may potentially collide with the robot in the current operating state. However, this feature is essentially based on calculating the potential risk trigger time according to the relative distance and relative approach rate. Dynamic obstacles typically exhibit random motion characteristics, and their direction of travel may deviate or change at any time. Therefore, further analysis of the motion direction relationship between the dynamic obstacle and the robot is needed.

[0065] Based on the above analysis, in some embodiments of the present invention, within a preset time window, the directional convergence trend between the dynamic obstacle and the robot and the stability of the dynamic obstacle's motion direction are analyzed according to the motion direction. Specifically, within the preset time window, based on the motion direction (motion direction vector) of the dynamic obstacle and the motion direction (motion direction vector) of the robot, the angle between the motion direction of the dynamic obstacle and the robot at each sampling moment is calculated; based on the angle between the motion direction, the changing trend of the angle between the motion direction at each sampling moment is analyzed to obtain the directional convergence trend between the dynamic obstacle and the robot within the preset time window. A more specific implementation is as follows: the ratio between the angle between the motion direction at any sampling moment and the angle between the motion direction at the next sampling moment is calculated to obtain the changing trend of the angle between the motion direction at each sampling moment, and all sampling moments within the preset time window are traversed, and the average value of the ratio between the motion direction angles is calculated to obtain the directional convergence trend between the dynamic obstacle and the robot within the preset time window. Furthermore, within a preset time window, based on the movement direction of the dynamic obstacle, the change in the movement direction of the dynamic obstacle at adjacent sampling times is obtained. That is, the angle difference between the movement direction of the dynamic obstacle at the current sampling time and the previous sampling time is calculated, and the normalized result of the angle difference is recorded as the change in the movement direction of the dynamic obstacle at the current sampling time. Then, the standard deviation of the change in the movement direction corresponding to all sampling times within the preset time window is calculated to obtain the stability of the movement direction of the dynamic obstacle within the preset time window.

[0066] Finally, based on the dynamic approach time margin between the dynamic obstacle and the robot within the preset time window, the directional convergence trend between the dynamic obstacle and the robot, and the stability of the dynamic obstacle's motion direction, the dynamic path convergence factor between the dynamic obstacle and the robot within the preset time window is determined. The formula for calculating the dynamic path convergence factor between the dynamic obstacle and the robot within the preset time window is as follows:

[0067]

[0068] In the formula, This represents the dynamic path convergence factor between dynamic obstacles and the robot within a preset time window. This indicates the dynamic approach time margin between dynamic obstacles and the robot within a preset time window; Represents the hyperbolic tangent function; Indicates the number of times within the preset time window The angle between the dynamic obstacle and the robot at each sampling time; Indicates the number of times within the preset time window The angle between the dynamic obstacle and the robot at each sampling time; This represents the total number of sampling moments within the preset time window; Indicates the number of times within the preset time window The change in the direction of motion of the dynamic obstacle at each sampling time; This represents the function for calculating the standard deviation. and Both represent denominator correction parameters, with the same dimensions as the denominator, and their values ​​are local minima greater than 0. and This is to prevent the denominator from being 0; for example, it can be set... , This can prevent the denominator from being 0 without significantly interfering with the calculation of normal values. Represented by natural constant An exponential function with base 0; This represents a linear normalization function, such as a max-min normalization function, used to normalize the dynamic path convergence factor to... Within the range, the maximum value in the maximum-minimum normalization function is the maximum allowable path convergence energy level based on historical calibration.

[0069] This indicates the increasing trend of directional convergence between dynamic obstacles and the robot within a preset time window. A larger ratio indicates a more pronounced deflection of the dynamic obstacle towards the robot, thus increasing the likelihood of dynamic path convergence between the obstacle and the robot; Dynamic Approach Time Margin The smaller the value, the shorter the time required for a dynamic obstacle to enter the robot's safety impact zone in the current operating state, and the greater the possibility of dynamic path convergence between the dynamic obstacle and the robot; This indicates the stability of the dynamic obstacle's movement direction within a preset time window. A smaller value indicates less directional fluctuation in the dynamic obstacle during its movement. The greater the reliability of the trend of directional convergence, the more reliable it becomes. right Weighted; The larger the value, the higher the tendency for paths to converge and collide between dynamic obstacles and the robot, that is, the larger the dynamic path convergence factor.

[0070] By analyzing the temporal correlation between the approach trend of dynamic obstacles and the robot's own motion state, the electronic control system can accurately identify the critical time interval for potential risks to occur before the emergency stop is triggered, thereby avoiding static judgment based solely on a single distance threshold.

[0071] The dynamic path convergence factor obtained above can express potential collision risks, but relying solely on risk assessment cannot directly guarantee the determinism of emergency stop execution. Emergency stop signals need to be transmitted in the control link and take a certain amount of time to execute the emergency stop response.

[0072] Therefore, after obtaining the dynamic path convergence factor, it is necessary to further combine it with the emergency stop trigger delay time to obtain the emergency stop timing offset index between the dynamic obstacle and the robot. This further includes: recording the time difference between the emergency stop trigger moment and the start of the emergency stop braking action as the emergency stop trigger delay time, and calculating the average emergency stop trigger delay time corresponding to multiple historical emergency stop triggers of the robot; based on the dynamic path convergence factor and the average emergency stop trigger delay time, obtaining the emergency stop timing offset index between the dynamic obstacle and the robot. Specifically, the formula for calculating the emergency stop timing offset index between the dynamic obstacle and the robot within a preset time window is:

[0073]

[0074] In the formula, This represents the emergency stop timing offset index between dynamic obstacles and the robot within a preset time window; This represents the dynamic path convergence factor between dynamic obstacles and the robot within a preset time window. Representing the history of robots The delay time for triggering an emergency stop; This represents the total number of times the robot has triggered an emergency stop in its history (this could be all the previous times, or the most recent 10 or 20 times). This represents the mean function; This represents a linear normalization function, such as a max-min normalization function, used to normalize the emergency stop timing offset exponent to... Within the range, the maximum value in the maximum-minimum normalization function is the maximum permissible emergency stop timing offset based on historical calibration.

[0075] This represents the average emergency stop trigger delay time for multiple emergency stop triggers in the robot's history. The average emergency stop trigger delay time is used to evaluate the robot's overall emergency stop execution performance. The larger the value, the higher the risk of convergence between dynamic obstacles and robot path within the preset time window, and the more obvious the execution after a sudden stop, that is, the larger the sudden stop timing offset index within the preset time window.

[0076] S300: Based on position data, analyze the fluctuation of the robot's running acceleration during the process from emergency stop triggering to emergency stop completion, and analyze the consistency of the execution time of multiple historical emergency stop triggering to emergency stop completion to obtain the robot's emergency stop execution consistency deviation coefficient.

[0077] The emergency stop timing offset index reflects the risk of emergency stop execution in the time dimension of the system. However, the average delay index alone cannot reflect the spatial or amplitude consistency of the execution action. That is, even if the average delay is small, if the execution unit responds unstablely under multiple instruction triggers, it will lead to inconsistent braking effects, and the safety risk still exists.

[0078] Based on the above analysis, in the embodiments of the present invention, the fluctuation of the robot's running acceleration during the process from emergency stop triggering to emergency stop completion is analyzed according to the position data, and the consistency of the execution time from multiple historical emergency stop triggering to emergency stop completion is analyzed to obtain the robot's emergency stop execution consistency deviation coefficient. Wherein:

[0079] The analysis of the fluctuations in robot acceleration during the emergency stop triggering and completion process further includes: First, the time difference between the emergency stop triggering time and the emergency stop completion time is recorded as the emergency stop operation execution time, and the emergency stop operation execution times corresponding to multiple historical emergency stop triggers are obtained; then, based on position data (including the robot's position data after multiple historical emergency stop triggers), the robot's running speed at each sampling moment during the emergency stop operation execution time is obtained, and the difference between the running speed at each sampling moment and the running speed at the previous sampling moment is calculated. Combined with the time interval between two sampling moments, the robot's running acceleration at each sampling moment during the emergency stop operation execution time is obtained; then, the standard deviation of all running accelerations is calculated to obtain the fluctuations in robot running acceleration during the emergency stop triggering and completion process. To more realistically reflect the fluctuations in robot running acceleration, the fluctuations in running acceleration corresponding to multiple historical emergency stops of the robot are obtained, and the mean of the fluctuations is calculated.

[0080] The analysis of the consistency in execution time between multiple historical emergency stop triggers and their completion includes: calculating the difference between the execution times of any two historical emergency stop triggers; averaging the differences between any two historical emergency stop triggers to obtain the consistency in execution time between multiple historical emergency stop triggers and their completion. It should be noted that the execution time of the emergency stop operation is related to the robot's performance and operating speed. However, patrolling robots in fixed scenarios are theoretically set to the same or similar operating speeds. Therefore, for this robot, the execution time of the emergency stop operation corresponding to multiple historical emergency stop triggers ignores the influence of operating speed differences and is assumed to be only related to the robot's emergency stop performance stability. That is, the average difference between any two emergency stop triggers can reflect the consistency in execution time between multiple historical emergency stop triggers and their completion.

[0081] Based on the fluctuations in robot acceleration and the consistency of execution time, the robot's emergency stop execution consistency deviation coefficient is obtained within a preset time window. Specifically, the formula for calculating the robot's emergency stop execution consistency deviation coefficient is as follows:

[0082]

[0083] In the formula, This represents the consistency deviation coefficient of the robot's emergency stop execution; This represents the average fluctuation of the robot's acceleration during all historical emergency stops from triggering to completion. Indicates the historical number The emergency stop triggers the corresponding emergency stop operation execution time; Indicates the historical number The emergency stop triggers the corresponding emergency stop operation execution time; This indicates the number of groups formed by any two emergency stop triggers in the robot's history (the history of multiple emergency stop triggers can be all the history of multiple emergency stop triggers, or the most recent 10 or 20 history of multiple emergency stop triggers). Indicates taking the absolute value; This represents a linear normalization function, such as a max-min normalization function, used to normalize the consistency deviation coefficient of emergency stop execution to... Within the range, the maximum value in the maximum-minimum normalization function is the maximum permissible consistency deviation of emergency stop execution based on historical calibration.

[0084] The standard deviation indicates the speed stability of the robot's emergency stop action. The smaller the standard deviation, the smoother the speed change and the more consistent the braking execution. Conversely, the larger the standard deviation, the more volatile the action and the unstable the execution. This represents the average time difference between any two emergency stop operations, reflecting the consistency in execution time among multiple emergency stop triggers. The smaller the difference, the more consistent the multiple emergency stop actions are. The larger the value, the greater the fluctuation in the emergency stop execution action and the poorer the consistency between different triggers. In other words, the lower the reliability of the robot's emergency stop, the higher the consistency deviation coefficient of the robot's emergency stop execution during the emergency stop operation.

[0085] S400: Determine the robot's emergency stop response reliability index based on the emergency stop timing offset index and the emergency stop execution consistency deviation coefficient.

[0086] The robot's emergency stop response reliability index is determined based on the emergency stop timing offset index and the emergency stop execution consistency deviation coefficient. Specifically, the formula for calculating the robot's emergency stop response reliability index within a preset time window is as follows:

[0087]

[0088] In the formula, This represents the reliability index of the robot's emergency stop response within a preset time window; This represents the consistency deviation coefficient of the robot's emergency stop execution; This represents the emergency stop timing offset index between dynamic obstacles and the robot within a preset time window; This represents the denominator correction parameter, which has the same dimensions as the denominator and takes the smallest value greater than 0. This is to prevent the denominator from being 0; for example, it can be set... This can prevent the denominator from being 0 without significantly interfering with the calculation of normal values. This represents a linear normalization function, such as a max-min normalization function, used to normalize the emergency stop response reliability index to... Within the range, the maximum value in the maximum-minimum normalization function is the preset upper limit of the range, which is set according to the maximum emergency stop response reliability level that appears in the historical operating data.

[0089] Multiplying the emergency stop timing deviation index by the execution consistency deviation coefficient yields the emergency stop response reliability index, which is used to evaluate the overall reliability level of emergency stop triggering and execution within a preset time window. This reliability index reflects both the response timing delay of the emergency stop command and the stability of the execution action, and can provide a more unified risk assessment basis for dynamic obstacle perception emergency stop electronic control systems.

[0090] By observing the response status of the emergency stop trigger command in the electronic control link, potential lag or failure risks in the emergency stop execution process can be effectively identified, thereby effectively improving the determinism of emergency stop triggering and the consistency of execution of the robot in complex dynamic environments, and also providing certain data support for subsequent analysis.

[0091] S500: Based on the emergency stop response reliability index and the available distance between the robot and the dynamic obstacle at the current moment, determine the emergency stop trigger adjustment coefficient of the robot at the current moment and compensate for the target execution gain.

[0092] Since there is a certain amount of time between triggering and completing an emergency stop (emergency stop operation execution time), the robot is in a decelerating forward running state during this period. Even with the robot's maximum power output (strongest braking force), it still needs to travel a certain distance. Therefore, there is a minimum theoretical distance for the robot's emergency stop process (assuming the robot's normal operating speed is constant, the minimum distance for an emergency stop when the robot's maximum power output is applied). This minimum theoretical distance can be obtained through preliminary experiments and periodically checked and corrected. If there is sufficient usable distance between the robot and dynamic obstacles when the emergency stop is triggered, a smaller braking force can be output, ensuring a safe emergency stop while also improving the robot's emergency stop stability and lifespan. Conversely, if the usable distance between the robot and dynamic obstacles is small when the emergency stop is triggered, a larger braking force is required to ensure the emergency stop is completed.

[0093] Based on the above analysis, in the embodiments of the present invention, the emergency stop triggering adjustment coefficient of the robot at the current moment is determined according to the emergency stop response reliability index and combined with the available distance between the robot and the dynamic obstacle at the current moment, and the target execution gain is compensated. Further, it includes:

[0094] First, obtain the minimum theoretical distance required from triggering an emergency stop to completing the emergency stop. Specifically, the minimum theoretical distance can be obtained through preliminary experiments, that is, by setting a fixed running speed when the robot is normally patrolling, and experimentally obtaining the minimum distance for the robot to stop when it is under maximum forced power output, and saving it to the robot performance parameter card. This can be obtained directly here.

[0095] Then, based on the position data, the available distance between the robot and the dynamic obstacle at the current moment is obtained. Specifically, the Euclidean distance between the robot's current position coordinates and the dynamic obstacle's current position coordinates is calculated; the maximum theoretical distance and maximum theoretical time required for an emergency stop to be triggered and completed are obtained. The maximum theoretical distance and maximum theoretical time can be obtained through prior experiments, i.e., setting a fixed running speed during normal robot patrol and experimentally obtaining the maximum distance and maximum time for emergency stop when the robot outputs minimum braking force; and obtaining the farthest forward distance of the dynamic obstacle within the maximum theoretical time, i.e., first extracting the fastest running speed of the dynamic obstacle (the maximum data among the dynamic obstacle running speeds already perceived by the robot), and obtaining the farthest forward distance of the dynamic obstacle by multiplying the maximum theoretical time by the fastest running speed; subtracting the maximum theoretical distance and the farthest forward distance from the Euclidean distance in turn, the available distance between the robot and the dynamic obstacle at the current moment is obtained.

[0096] Then, the ratio between the available distance and the minimum theoretical distance is calculated to obtain the robot's emergency stop safety margin at the current moment.

[0097] Next, based on the emergency stop response reliability index and the emergency stop safety margin, the robot's emergency stop trigger adjustment coefficient for the current moment is determined. Specifically, the formula for calculating the robot's emergency stop trigger adjustment coefficient for the current moment is:

[0098]

[0099] In the formula, This represents the robot's emergency stop trigger adjustment coefficient at the current moment; This represents the reliability index of the robot's emergency stop response within a preset time window; This indicates the available distance between the robot and the dynamic obstacle at the current moment; This represents the minimum theoretical distance required for the robot to complete an emergency stop after it is triggered. Represented by natural constant An exponential function with base 0.

[0100] This represents the robot's emergency stop safety margin at the current moment. The larger this ratio, the more available distance there is, resulting in a higher emergency stop safety margin and a smaller emergency stop trigger adjustment coefficient for the robot. Conversely, a smaller emergency stop safety margin may indicate insufficient or dangerous conditions, leading to a larger emergency stop trigger adjustment coefficient for the robot. When this occurs, it indicates that the available distance between the dynamic obstacle and the robot is less than or equal to the minimum theoretical distance. Therefore, the maximum emergency stop trigger adjustment coefficient can be directly set, i.e. Emergency stop response reliability index This reflects the overall reliability level of emergency stop triggering and execution. The larger the value, the smaller the corresponding emergency stop triggering adjustment coefficient of the robot. The smaller the value, the lower the closed-loop reliability of the electronic control system, and the more necessary it is to enhance and adjust it, i.e., to improve the emergency stop response.

[0101] Finally, the target execution gain is compensated based on the robot's emergency stop trigger adjustment coefficient at the current moment. Specifically, the target execution gain refers to the control parameter output by the electronic control system to the drive execution unit, used to adjust the intensity and speed of the drive execution unit's response to the emergency stop action. The maximum and minimum values ​​of the allowable target execution gain parameter output by the electronic control system to the emergency stop execution unit are obtained, and are denoted as follows: Then, combining the emergency stop trigger adjustment coefficient, the compensated target execution gain is calculated as follows:

[0102]

[0103] In the formula, Indicates the target execution gain after compensation; This represents the minimum value of the target execution gain parameter that the electronic control system outputs to the emergency stop execution unit; This indicates the maximum value of the target execution gain parameter allowed by the electronic control system output to the emergency stop execution unit; This represents the robot's emergency stop trigger adjustment coefficient at the current moment.

[0104] After obtaining the compensated target execution gain, during the dynamic obstacle perception emergency stop electronic control process, the compensated target execution gain is sent to the drive execution unit along with the current emergency stop command. The drive execution unit adjusts the braking force according to the target execution gain, including power output or braking force.

[0105] During the emergency stop, the electronic control unit collects monitoring data (including motion speed data, emergency stop trigger delay time, emergency stop operation execution time, etc.) through sensors or drive execution units, and then updates the closed-loop state to provide a data basis for the next control cycle. Based on the data, the emergency stop trigger adjustment coefficient is recalculated. By dynamically adjusting the emergency stop trigger adjustment coefficient, continuous closed-loop adaptive adjustment is achieved to ensure that the patrol robot's actions during obstacle perception and emergency stop are fast, safe, and stable in different dynamic environments.

[0106] It should be noted that, in the embodiments of the present invention, the emergency stop operation or braking execution includes not only the highest priority emergency stop, but also obstacle avoidance braking, deceleration and stopping triggered by dynamic obstacles during the robot's patrol, as well as periodic braking test behaviors. Since these behaviors reuse the same underlying drive and communication link, their timing response characteristics are consistent with those of the emergency stop.

[0107] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A dynamic obstacle perception and emergency stop control method for safe robot patrolling, characterized in that, The method includes: The position data of the dynamic obstacles and the robot at each moment during the movement are collected separately; Based on the location data, the dynamic path convergence factor between the dynamic obstacle and the robot is analyzed, and combined with the emergency stop trigger delay time, the emergency stop timing offset index between the dynamic obstacle and the robot is obtained. Based on the location data, the fluctuation of the robot's running acceleration during the process from emergency stop triggering to emergency stop completion is analyzed, and the consistency of the execution time from multiple historical emergency stop triggering to emergency stop completion is analyzed to obtain the robot's emergency stop execution consistency deviation coefficient. The robot's emergency stop response reliability index is determined based on the emergency stop timing offset index and the emergency stop execution consistency deviation coefficient. Based on the emergency stop response reliability index and the available distance between the robot and the dynamic obstacle at the current moment, the emergency stop trigger adjustment coefficient of the robot at the current moment is determined to compensate for the target execution gain. Based on the location data, analyze the dynamic path convergence factor between dynamic obstacles and the robot, including: Preset time window; Within a preset time window, based on the location data, the relative distance between the dynamic obstacle and the robot is obtained, as well as the movement direction of the dynamic obstacle and the robot respectively. Within a preset time window, the dynamic approach time margin between the dynamic obstacle and the robot is analyzed based on the relative distance. Within a preset time window, based on the direction of motion, the directional convergence trend between the dynamic obstacle and the robot is analyzed, and the stability of the direction of motion of the dynamic obstacle is analyzed. Combined with the dynamic approach time margin, the dynamic path convergence factor between the dynamic obstacle and the robot within the preset time window is determined. By combining the emergency stop trigger delay time, the emergency stop timing offset index between the dynamic obstacle and the robot is obtained, including: The time difference between the emergency stop triggering time and the start of the emergency stop braking action is recorded as the emergency stop triggering delay time, and the average emergency stop triggering delay time corresponding to multiple historical emergency stop triggers is calculated. Based on the dynamic path convergence factor and the average emergency stop trigger delay time, the emergency stop timing offset index between the dynamic obstacle and the robot is obtained. Based on the location data, analyze the fluctuations in the robot's acceleration during the process from triggering the emergency stop to completing the emergency stop, including: The time difference between the emergency stop trigger time and the emergency stop completion time is recorded as the emergency stop operation execution time. Based on the location data, the robot's running acceleration at each sampling moment during the emergency stop operation execution time is obtained, the standard deviation of all the running accelerations is calculated, and the fluctuation of the robot's running acceleration during the process from the triggering of the emergency stop to the completion of the emergency stop is obtained; Analyze the consistency of execution time from triggering to completing multiple emergency stops in history, including: Calculate the difference between the execution times of any two emergency stop triggers in history, iterate through all the difference values ​​corresponding to two emergency stop triggers in history, calculate the average value, and obtain the consistency in execution time from multiple emergency stop triggers to the completion of the emergency stop in history.

2. The dynamic obstacle perception and emergency stop control method for safe robot patrolling according to claim 1, characterized in that, Within a preset time window, based on the relative distance, the dynamic approach time margin between the dynamic obstacle and the robot is analyzed, including: Within a preset time window, the approach rate between the dynamic obstacle and the robot is obtained based on the relative distance; Within a preset time window, based on the approach rate, the rate of change of the approach rate between the dynamic obstacle and the robot at adjacent sampling times is analyzed. Combined with the average approach rate within the preset time window, the dynamic approach time margin between the dynamic obstacle and the robot is obtained.

3. The dynamic obstacle perception and emergency stop electronic control method for safe robot patrolling according to claim 2, characterized in that, Within a preset time window, based on the direction of motion, the convergence trend between the dynamic obstacle and the robot is analyzed, and the stability of the dynamic obstacle's motion direction is analyzed, including: Within a preset time window, the angle between the dynamic obstacle and the robot is calculated based on the direction of motion. The changing trend of the angle of motion at each sampling moment is analyzed to obtain the directional convergence trend between the dynamic obstacle and the robot within the preset time window. Within a preset time window, based on the motion direction of the dynamic obstacle, the change in the motion direction of the dynamic obstacle at adjacent sampling times is obtained, thus obtaining the stability of the motion direction of the dynamic obstacle.

4. The dynamic obstacle perception and emergency stop electronic control method for safe robot patrolling according to claim 3, characterized in that, Analyzing the changing trend of the angle between the motion directions at each sampling moment, the convergence trend of the dynamic obstacles and the robot within a preset time window is obtained, including: Calculate the ratio between the angle of motion direction at any sampling time and the angle of motion direction at the next sampling time to obtain the trend of change of the angle of motion direction at each sampling time; By iterating through all sampling moments within a preset time window, the average value of the ratios between the angles of the motion directions is calculated to obtain the directional convergence trend between the dynamic obstacle and the robot within the preset time window.

5. The dynamic obstacle perception and emergency stop electronic control method for safe robot patrolling according to claim 1, characterized in that, Based on the emergency stop response reliability index and the available distance between the robot and the dynamic obstacle at the current moment, the emergency stop trigger adjustment coefficient of the robot at the current moment is determined, including: Obtain the minimum theoretical distance required from the current emergency stop trigger to the completion of the emergency stop; Based on the location data, the available distance between the robot and the dynamic obstacle at the current moment is obtained; Calculate the ratio between the available distance and the minimum theoretical distance to obtain the robot's emergency stop safety margin at the current moment; Based on the emergency stop response reliability index and the emergency stop safety margin, the emergency stop trigger adjustment coefficient of the robot at the current moment is determined.

6. The dynamic obstacle perception and emergency stop electronic control method for safe robot patrolling according to claim 5, characterized in that, Compensation is provided for the target execution gain, including: The maximum and minimum values ​​of the target execution gain parameters allowed by the electronic control system output to the emergency stop execution unit are obtained, and the target execution gain is compensated in combination with the emergency stop trigger adjustment coefficient.

Images