Active obstacle avoidance control method and system for a catering service robot

Abstract

This invention relates to the field of robot obstacle avoidance technology, specifically to an active obstacle avoidance control method and system for a catering service robot. The method determines the robot's current food delivery stage, acquires basic spatial demand characteristics, and analyzes the robot's sway sensitivity coefficient to obtain the robot's current stable passage demand. It also analyzes the obstacle's forward, lateral, and docking pressure on the robot to determine the degree of intrusion pressure. Furthermore, it detects the unobstructed space in the forward, lateral, and table-side docking areas to determine the current spatial support level. Based on the spatial support level, stable passage demand, and intrusion pressure level, the method obtains the robot's current safety status maintenance value and outputs the robot's current obstacle avoidance control strategy. This invention, by integrating demand, pressure, and support aspects, outputs a more reasonable and executable active obstacle avoidance control strategy, improving the robot's stability and efficiency.

Description

Technical Field

[0001] This invention relates to the field of robot obstacle avoidance technology, specifically to an active obstacle avoidance control method and system for a catering service robot. Background Technology

[0002] With the development of technology, intelligent robots have come into people's view. In order to improve the efficiency of restaurants and effectively reduce their operating costs, catering service robots are becoming increasingly popular due to their lower cost compared to human labor and their high-tech positioning.

[0003] However, during the process of serving food, walking, and stopping at tables, food service robots inevitably encounter situations such as customers getting up, pulling out chairs, waiters crossing the road, children suddenly running out, and items falling to the ground. Therefore, the active obstacle avoidance of food service robots is extremely challenging.

[0004] Current active obstacle avoidance control for food service robots typically revolves around obstacle detection results. That is, upon detecting obstacles such as people or objects ahead, the robot directly decelerates, stops, or detours. While this approach can handle general collision risks, its implementation logic primarily focuses on "whether there is an obstacle ahead," with little analysis of the required safety space for the robot in different delivery scenarios, or how sudden intrusions alter this safety space. For example, when leaving the food preparation area, the robot needs more forward clearance; when delivering food through the main aisle, it needs continuous passage; and when delivering to a table, it needs single-sided stopping space. However, situations such as customers getting up, pulling out chairs, waiters crossing the path, children suddenly darting out, or items falling to the ground can cause the required space to shorten forward, be compressed laterally, or have its stopping area interrupted. If a fixed safety distance or a direct response after a single detection is still used for different delivery scenarios, it is difficult to reflect the differences in data performance caused by different scenarios, and it is also difficult to quantify the actual impact of intrusions on the current delivery action. In addition, the food delivery robot will exhibit swaying differences when adjusting its path while carrying food. If controlled according to the logic of ordinary mobile robots, it is easy to cause frequent sudden stops, excessive detours, or a decrease in the stability of the dishes. Summary of the Invention

[0005] In order to solve the above technical problems, the purpose of this invention is to provide an active obstacle avoidance control method and system for catering service robots.

[0006] This invention provides an active obstacle avoidance control method for a catering service robot, comprising: Determine the current food delivery stage of the robot and obtain the basic spatial requirement characteristics of the current food delivery stage; Collect the robot's three-axis acceleration data, analyze the robot's current sway amplitude and recovery time, and obtain the robot's current sway sensitivity coefficient; By combining the basic spatial demand characteristic value and the sway sensitivity coefficient, the current stable passage demand of the robot is obtained; The system detects the position, speed, and occupancy data of obstacles, and analyzes the forward pressure, lateral pressure, and docking pressure of obstacles on the robot to obtain the degree of intrusion and pressure of obstacles on the robot. The system detects the continuous barrier-free distance along the tangent direction of the current planned trajectory and the barrier-free space span on both sides along the normal direction, as well as the barrier-free space area of ​​the table-side resting area, to determine the current level of spatial support. Based on the spatial support level, the stable passage requirement level, and the intrusion pressure level, the robot's current safety status maintenance value is obtained, and the robot's current obstacle avoidance control strategy is output.

[0007] Furthermore, determining the current stage of the robot's food delivery process includes: Divide the restaurant into different scene areas on a static map; After receiving the food delivery task, the robot obtains the location of the target table and the robot's current location. It then matches the current location with the restaurant's static map to determine the scene area where the robot is currently located. Combining the task progress and the target table location, it determines the current stage of the robot's food delivery.

[0008] Furthermore, the food delivery stage includes the food pickup stage, the departure stage, the passage stage, and the docking stage; the scene area includes the food dispensing area, the main passage area, the corner passing area, the table-side docking area, and the return passage area. The robot's current location is then matched with the restaurant's static map to determine its current scene area. Combined with the task progress and the target table's location, the robot's current food delivery stage is determined, including: If the robot is currently in the food dispensing area and has not yet picked up any food, then the robot is currently in the food picking phase. If the robot is currently in the food dispensing area and has already picked up a meal, then the robot is considered to be in the departure phase. If the robot is currently located in the main passage area, corner meeting area or return passage area, and is moving towards the target table or base station, then the robot is determined to be in the passage phase. If the robot's current position is in the tableside docking area and the distance to the target table position is less than the preset distance, then the robot is determined to be in the docking phase.

[0009] Furthermore, the robot's three-axis acceleration data is collected, and the robot's current sway amplitude and recovery time are analyzed to obtain the robot's current sway sensitivity coefficient, including: Collect the robot's three-axis acceleration data, calculate the root mean square of the three-axis acceleration, and obtain the robot's current sway amplitude; The time required for the sway amplitude to decrease to below 10% of the initial maximum sway amplitude is monitored to obtain the robot's current stabilization time. Based on the sway amplitude and the stabilization time, combined with the maximum permissible sway amplitude and maximum permissible stabilization time at the factory calibration, and by introducing sway weight and stabilization weight, the robot's current sway sensitivity coefficient is obtained.

[0010] Furthermore, by combining the basic spatial demand characteristic value and the sway sensitivity coefficient, the robot's current stable passage demand is obtained, including: Based on the basic space demand characteristic value and the sway sensitivity coefficient, combined with the maximum basic space demand characteristic value and the maximum sway sensitivity coefficient, and by introducing space demand weight and sway sensitivity weight, the current stable passage demand of the robot is obtained.

[0011] Furthermore, by detecting the position, velocity, and occupancy data of obstacles, and analyzing the forward pressure, lateral pressure, and docking pressure sub-items of the obstacles on the robot, the degree of intrusion and pressure exerted by the obstacles on the robot is obtained, including: The system detects the position and speed of obstacles and their occupancy data in the docking area at the target table location. Combined with the robot's current position and running speed, it obtains the longitudinal distance, longitudinal approach speed, lateral distance, and lateral approach speed between the obstacle and the robot in the robot's direction of travel. Based on the longitudinal distance and the longitudinal approach speed, the forward pressure sub-item of the obstacle on the robot is obtained; Based on the lateral distance and the lateral approach speed, the lateral pressure sub-item of the obstacle on the robot is obtained; Based on the data on the occupancy of the obstacle in the docking area at the target table location, and combined with the maximum allowable occupancy of the docking area at the target table location, the sub-item of the obstacle's docking pressure on the robot is obtained; Based on the forward pressure sub-item, the lateral pressure sub-item, and the docking pressure sub-item, and by introducing an intrusion type weight, the degree of intrusion pressure exerted by the obstacle on the robot is obtained.

[0012] Furthermore, the continuous barrier-free distance along the tangent direction of the current planned trajectory and the barrier-free space span on both sides along the normal direction are detected, as well as the area of ​​the barrier-free space in the table-side resting area, to determine the current level of spatial support, including: The system detects the continuous barrier-free distance along the tangent of the current planned trajectory and the barrier-free space span on both sides along the normal direction, as well as the area of ​​the barrier-free space in the tableside resting area. The forward space adequacy is obtained by combining the continuous unobstructed distance in the tangential direction with the standard unobstructed distance requirement in the tangential direction. Based on the unobstructed space span on both sides of the normal direction, and combined with the unobstructed standard requirement space span on both sides of the normal direction, the lateral space adequacy is obtained. Based on the area of ​​the accessible space in the tableside parking area, and combined with the area of ​​the accessible standard requirements for the tableside parking area, the sufficiency of the tableside parking space is obtained. Based on the aforementioned forward space adequacy, lateral space adequacy, and tableside docking space adequacy, and by introducing scene weighting coefficients, the current level of spatial support is determined.

[0013] Furthermore, based on the spatial support level, the stable passage requirement, and the intrusion pressure level, the robot's current safety status maintenance value is obtained, and the robot's current obstacle avoidance control strategy is output, including: Based on the spatial support level, the stable passage requirement level, and the intrusion pressure level, the current safety status maintenance value of the robot is obtained; A first security state threshold and a second security state threshold are preset; the first security state threshold is greater than the second security state threshold. If the safety status maintenance value is greater than or equal to the first safety status threshold, the robot is currently in a safe state and is proceeding normally. If the safety status maintenance value is less than the first safety status threshold and greater than or equal to the second safety status threshold, the robot is currently in a warning state. Then, based on the forward space adequacy, the lateral space adequacy, the tableside docking space adequacy, and the performance of the forward pressure sub-item, the lateral pressure sub-item, and the docking pressure sub-item, an obstacle avoidance strategy is formulated. If the safety status maintenance value is less than the second safety status threshold, the robot is currently in a dangerous state and an emergency stop strategy is executed.

[0014] According to a second aspect of the present invention, an active obstacle avoidance control system for a catering service robot is provided, comprising: a memory and a processor, wherein: The memory is used to store program code; The processor is configured to read program code stored in the memory and execute the method described in the first aspect of the present invention.

[0015] Furthermore, the processor includes: The stable passage demand analysis module is used to determine the current food delivery stage of the robot and obtain the basic space demand characteristic value of the current food delivery stage; and to collect the robot's three-axis acceleration data, analyze the robot's current swaying amplitude and recovery time, and obtain the robot's current swaying sensitivity coefficient; and then, by combining the basic space demand characteristic value and the swaying sensitivity coefficient, obtain the robot's current stable passage demand degree. The intrusion and pressure analysis module is used to detect the position, speed, and occupancy data of obstacles, and analyze the forward pressure, lateral pressure, and docking pressure of obstacles on the robot to obtain the degree of intrusion and pressure of obstacles on the robot. The spatial support analysis module is used to detect the continuous barrier-free distance in the tangent direction of the current planned trajectory and the barrier-free space span on both sides in the normal direction, as well as the area of ​​the barrier-free space in the table-side resting area, to determine the current spatial support level. The obstacle avoidance control module is used to obtain the robot's current safety status maintenance value based on the spatial support level, the stable passage requirement level, and the intrusion pressure level, and output the robot's current obstacle avoidance control strategy.

[0016] Compared to existing technologies, the active obstacle avoidance control method and system for catering service robots provided by this invention have the following beneficial effects: By determining the current food delivery stage of the robot and obtaining the basic spatial demand characteristic values ​​for that stage (the basic spatial demand characteristic values ​​are fundamental data on the demand side, reflecting the inherent spatial demands in restaurant operation scenarios), and by using the sway sensitivity coefficient as a correction data on the demand side, reflecting the additional requirements for spatial stability under the food-carrying state, the combination of these two factors yields a precise demand level. Furthermore, by detecting the position, speed, and occupancy data of obstacles, the degree of intrusion and pressure exerted by obstacles on the robot is obtained. Finally, by quantifying the pressure mechanisms of different intrusion behaviors, a weighted average is achieved. The comprehensive impact assessment of intrusion avoids the one-sidedness of single judgments. By detecting the continuous unobstructed distance in the tangential direction of the current planned trajectory and the unobstructed space span on both sides in the normal direction, as well as the area of ​​the unobstructed space in the table-side docking area, the current spatial support level is determined, solving the problem that a single distance cannot cover all scenario requirements. By considering the compression performance of the overall available space, obstacle avoidance decisions can be made earlier, thereby improving the passability of subsequent plans. In other words, by comprehensively considering the three aspects of demand, pressure, and support, the current safety level of the robot is comprehensively reflected, and the output active obstacle avoidance control strategy is more reasonable and executable, improving the stability and efficiency of the robot's movement. Attached Figure Description

[0017] Figure 1 This is a flowchart of an active obstacle avoidance control method for a catering service robot, provided as an embodiment of the present invention. Detailed Implementation

[0018] like Figure 1 As shown, an embodiment of the present invention provides an active obstacle avoidance control method for a catering service robot, specifically including: S100: Determine the current food delivery stage of the robot and obtain the basic space requirement characteristic value of the current food delivery stage.

[0019] Food service robots have different space requirements at different stages of food delivery. For example, when the robot is moving in the main aisle, it moves at a faster speed and has a higher demand for forward space. When the robot stops at the target table to prepare food, its range of movement is very small or it does not move at all, so it has a certain demand for surrounding space, but the demand for space is lower than when it is moving quickly.

[0020] Based on the above analysis, in the embodiments of the present invention, the current food delivery stage of the robot is determined, and the basic spatial requirement characteristic values ​​of the current food delivery stage are obtained. Further aspects include: First, different scene areas are divided on the restaurant's static map. Specifically, a high-precision point cloud map of the restaurant is constructed using laser SLAM (Simultaneous Localization and Mapping) technology, with a resolution set to 0.05m, to obtain the restaurant's static map. Through manual annotation and automatic completion via visual semantic segmentation, five scene areas are marked on the restaurant's static map: dining area, main aisle, corner passing area, table-side parking area, and return passage. Each scene area is assigned a label: C1 (dining area), C2 (main aisle), C3 (table-side parking area), C4 (corner passing area), and C5 (return passage). The physical attributes of each scene area are also recorded, including the width of the main aisle (default 1.5-2m), the distance between the table-side parking area and the table (default 0.3-0.5m), and the minimum passage radius of the corner passing area (default 1.2m).

[0021] Then, after receiving the delivery task, the robot obtains the target table location and its current location. It matches its current location with the restaurant's static map to determine the robot's current scene area. Combining the task progress and the target table location, it determines the current delivery stage. Specifically, the delivery stages include the pickup stage, departure stage, passage stage, and docking stage. After receiving the delivery task, the robot obtains the target table location corresponding to the target table number and its current location (output by the SLAM positioning module). It matches its current location with the restaurant's static map to determine the robot's current scene area (C1, C2, C3, C4, C5). Furthermore, it obtains the task progress, including the weight of the food to determine if it has been picked up, and the minimum distance between its current location and the target table location to determine if it has reached within 1 meter of the target table location. Then, combining the task progress (e.g., whether the food has been picked up, whether it has reached within 1 meter of the target table location), it determines the robot's current delivery stage according to predefined rules, where: If the robot is currently in the food dispensing area and has not yet picked up any food, then the robot is currently in the food picking phase. If the robot is currently in the food dispensing area and has already picked up a meal, then the robot is considered to be in the departure phase. If the robot is currently located in the main passage area, corner meeting area or return passage area, and is moving towards the target table or base station, then the robot is determined to be in the passage phase. If the robot's current position is in the tableside docking area and the distance to the target table position is less than the preset distance (1m), then the robot is determined to be in the docking stage.

[0022] In addition, preset basic space requirement characteristic values ​​for different food delivery stages are established. Specifically, the basic space requirement characteristic values ​​for each food delivery stage reflect the robot's fundamental dependence on safe space. The basic space requirement characteristic values ​​for the food retrieval stage are set based on empirical data or preliminary debugging data. (Due to slow movement speed and small activity range, the basic dependence on safe space is minimal), the characteristic value of basic space requirement during the departure phase is... (Slow movement speed but large activity range, therefore a high basic dependence on safe space), the basic space requirement characteristic value for the passage phase is: (Due to their high movement speed and large range of activity, their basic dependence on safe space is the highest) and the basic space requirement characteristics during the docking phase are: (The movement speed is slow and the range of activity is small, so the basic dependence on the safe space is relatively small).

[0023] S200: Collects three-axis acceleration data of the robot, analyzes the current sway amplitude and recovery time of the robot, and obtains the current sway sensitivity coefficient of the robot.

[0024] The robot exhibits varying degrees of shaking depending on its food-carrying state. Shaking is less when empty, more pronounced with solid food, and most intense with soup. Different shaking conditions elicit different levels of sensitivity to traffic adjustments. If the robot displays significant shaking or slow recovery after shaking during minor path adjustments (turning angle ≤ 5°) or normal movement (speed fluctuation ≤ 0.1 m / s), it indicates that the current food-carrying state is more sensitive to traffic adjustments, resulting in shaking feedback.

[0025] Based on the above analysis, in an embodiment of the present invention, by collecting the robot's three-axis acceleration data, analyzing the robot's current sway amplitude and recovery time, the robot's current sway sensitivity coefficient is obtained. Further aspects include: First, the robot's three-axis acceleration data is collected, and the root mean square (RMS) of the three-axis acceleration is calculated to obtain the robot's current sway amplitude. Specifically, a three-axis accelerometer (model: ADXL345, sampling frequency 50Hz) is installed on the bottom of the robot's tray. When the robot makes slight path adjustments (steering angle ≤ 5°) or moves normally (speed fluctuation ≤ 0.1m / s), the robot's three-axis acceleration data is continuously collected, and the three-axis acceleration data for the previous 10 seconds is obtained (if the total duration is less than 10 seconds, the three-axis acceleration data for the total duration is obtained). All data are aligned with a unified timestamp to ensure time consistency in subsequent analysis. The RMS value of the three-axis acceleration can quantify the sway amplitude; therefore, calculating the RMS of the three-axis acceleration yields the robot's current sway amplitude as follows: In the formula, This indicates the current amplitude of the robot's sway; Indicates the 10th second preceding the current time. A robot at each sampling point The acceleration of the shaft; Indicates the 10th second preceding the current time. A robot at each sampling point The acceleration of the shaft; Indicates the 10th second preceding the current time. A robot at each sampling point The acceleration of the axis (which should be understood) The acceleration of the axis is the acceleration after subtracting the acceleration due to gravity. This represents the total number of sampling points within 10 seconds prior to the current time.

[0026] Then, the time required for the sway amplitude to decrease to below 10% of the initial maximum sway amplitude is monitored to obtain the robot's current stabilization time. Specifically, the sway amplitudes corresponding to all moments prior to the current time are differentially analyzed. The time interval closest to the current time, where the sway amplitude continuously decreases (sway amplitude difference value is less than 0) and the sway amplitude decreases to 10% of the initial maximum sway amplitude, is obtained as the robot's current stabilization time.

[0027] Finally, based on the shaking amplitude and stabilization time, combined with the maximum allowable shaking amplitude and maximum allowable stabilization time specified in the factory calibration, and introducing shaking weights and stabilization weights, the robot's current shaking sensitivity coefficient is obtained. Specifically, the robot's shaking amplitude and stabilization time directly reflect the sensitivity of the robot's food-carrying state to path adjustments. The more severe the shaking and the slower the stabilization, the more unable the robot's current food-carrying state can withstand frequent path adjustments. Therefore, the formula for calculating the robot's shaking sensitivity coefficient is defined as follows: In the formula, This indicates the robot's current sway sensitivity coefficient; This indicates the current amplitude of the robot's sway; Indicates the maximum permissible shaking amplitude at the time of factory calibration; This indicates the robot's current stabilization time; Indicates the maximum permissible stabilization time at the factory calibration; Indicates the sway weight, Indicates the stabilization weight. , and The value can be set according to the current delivery stage. For example, if the current delivery stage is the passage stage or the departure stage, the travel speed or acceleration is relatively large, and the requirement for a large sway amplitude is high, then the value can be set accordingly. , The current delivery phase is either the pickup phase or the docking phase, requiring a quick return to stability; therefore, the following settings are configured. , .

[0028] It should be noted that if the robot has no accumulated acceleration data during the initial stage or within the past 10 seconds, it will obtain the type of food currently being served (soup, solid food, or empty) and apply the default sensitivity coefficient based on the type of food currently being served. For soup, the corresponding sensitivity coefficient is... Solid food products No-load correspondence If it includes both soup and solid food, then .

[0029] Thus, by measuring the amplitude of the robot's food-carrying sway and the recovery time, the sensitivity of different food-carrying states to traffic adjustments is quantified, providing a basis for correcting the intensity of basic requirements in the scenario and solving the problem that traditional models do not consider differences in food-carrying states.

[0030] S300: By combining the basic spatial demand characteristic value and the sway sensitivity coefficient, the current stable passage demand of the robot is obtained.

[0031] By performing a correlation analysis on the basic spatial demand characteristic value and the sway sensitivity coefficient, if the robot currently relies on continuous passage space and the sway feedback is obvious, it indicates that it is more dependent on stable forward space; if the robot currently relies on single-sided docking space and the recovery after swaying is slow, it indicates that it is more dependent on stable docking space; thus, the stability passage demand of the scene can be extracted.

[0032] Since the basic spatial demand characteristic value is the basic data on the demand side, reflecting the inherent spatial demand in the restaurant operation scenario; the sway sensitivity coefficient is the corrected data on the demand side, reflecting the additional requirements of the food-carrying state for spatial stability, the combination of the two can obtain an accurate demand level.

[0033] Based on the above analysis, in the embodiments of the present invention, the current stable passage requirement of the robot is obtained by combining the basic spatial requirement characteristic value and the sway sensitivity coefficient. Further aspects include: Based on the basic space demand characteristic value and the sway sensitivity coefficient, combined with the maximum basic space demand characteristic value and the maximum sway sensitivity coefficient, and introducing space demand weights and sway sensitivity weights, the robot's current stable passage demand is obtained. Specifically, space demand weights are set. Shaking sensitive weights This weight can be adjusted based on the restaurant's actual operational data. For example, if soup accounts for a high proportion of the menu, it can be adjusted to a weight based on space requirements. Shaking sensitive weights The maximum basic space requirement characteristic value is 0.9 (basic space requirement characteristic value during passage), and the maximum sway sensitivity coefficient is 2 (when both sway amplitude and recovery time reach twice the maximum allowable value). This is the maximum sway sensitivity coefficient value calibrated at the factory for the robot equipment, and it will not be exceeded in actual use. Therefore, the formula for calculating the robot's stable passage requirement is defined as follows: In the formula, This indicates the current stable traffic demand for the robot; This indicates the robot's current sway sensitivity coefficient; This represents the basic spatial requirement characteristic value of the robot at the current food delivery stage; This represents the characteristic value of the maximum basic space requirement; Indicates the maximum sway sensitivity coefficient; Indicates the weight of spatial demand; This indicates the weights that are sensitive to fluctuations.

[0034] Stable traffic demand This reflects the degree to which robots rely on safe spaces. The closer the value is to 1, the higher the requirement for spatial stability and continuity. Therefore, based on the stability requirement... Clearly define the core needs for obstacle avoidance, such as Priority should be given to ensuring the continuity and stability of the space.

[0035] S400: Detects the position, speed, and occupancy data of obstacles, and analyzes the forward pressure, lateral pressure, and docking pressure of obstacles on the robot to obtain the degree of intrusion and pressure of obstacles on the robot.

[0036] The aforementioned steps, having obtained a stable passage demand level from the robot's demand side, also require analysis of the intrusion behavior of potential obstacles along the passage path relative to the robot's current direction of travel. If an obstacle continuously approaches the robot's direction of travel, it indicates increased forward pressure; if an obstacle laterally cuts into the area the robot is expected to pass through, it indicates increased lateral pressure; if an obstacle is located in or continues to expand into the tableside stopping area, it indicates increased stopping pressure. Therefore, intrusion pressure characteristics can be extracted based on pressure changes. Specifically, forward and lateral pressure characteristics can be obtained by quantifying the relative positional changes (distance and approach speed) between the obstacle and the robot, and stopping pressure characteristics can be obtained by quantifying the expansion of the obstacle's occupation area in the tableside stopping area at the target table location, thus yielding the degree of intrusion pressure.

[0037] Based on the above analysis, in embodiments of the present invention, by detecting the position, speed, and occupancy data of obstacles, the forward pressure, lateral pressure, and docking pressure sub-items of the obstacles on the robot are analyzed respectively to obtain the degree of intrusion and pressure exerted by the obstacles on the robot. Further, this includes: First, the position, speed, and occupancy data of obstacles relative to the target table's edge docking area are detected. Combined with the robot's current position and speed, the longitudinal distance, longitudinal approach speed, lateral distance, and lateral approach speed between the obstacle and the robot in the robot's direction of travel are obtained. Specifically, visual sensors, LiDAR, and ultrasonic sensors are used to identify obstacles, detect their position, speed, and occupancy data relative to the target table's edge docking area (occupancy data is estimated through multiple sensor data), and extract the angle between the obstacle's movement direction and the robot's movement direction. Existing technologies are not detailed here. Based on the obstacle's position and velocity, the robot's position and speed, and the angle between the obstacle's and robot's directions of movement, calculate the straight-line distance between the obstacle and the robot. Then, calculate the product of the straight-line distance and the cosine of the supplementary angle to obtain the longitudinal distance between the obstacle and the robot in the robot's direction of travel. Calculate the product of the straight-line distance and the sine of the angle to obtain the lateral distance between the obstacle and the robot in the robot's direction of travel. Calculate the product of the obstacle's velocity and the cosine of the supplementary angle, and add it to the robot's speed to obtain the longitudinal approach velocity between the obstacle and the robot in the robot's direction of travel. Calculate the product of the obstacle's velocity and the sine of the angle to obtain the lateral approach velocity between the obstacle and the robot in the robot's direction of travel.

[0038] Then, based on the longitudinal distance and longitudinal approach velocity, the forward pressure sub-term of the obstacle on the robot is obtained. Specifically, the quantification formula for the forward pressure sub-term of the obstacle on the robot is: In the formula, This indicates the degree of forward pressure exerted by the obstacle on the robot, i.e., the forward pressure sub-item; Indicates the speed of the obstacle; Represents the cosine function; Indicates the angle between the direction of obstacle movement and the direction of robot movement; Indicates the robot's operating speed; Indicates the straight-line distance between the obstacle and the robot; This represents a linear normalization function, such as a max-min normalization function, used to normalize forward compression values ​​to... Within the range, the maximum and minimum values ​​in the maximum-minimum normalization function are determined based on the extreme values ​​of the forward compression values ​​in the historical monitoring database. When the normalized object is greater than the maximum value, the normalization result is limited to 1, and when the normalized object is less than the minimum value, the normalization result is limited to 0. 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.

[0039] This represents the longitudinal approach speed of the obstacle and the robot in the robot's direction of travel. The larger the value, the faster the two are approaching each other in the robot's direction of travel, and the greater the forward pressure exerted by the obstacle on the robot. This represents the longitudinal distance between the obstacle and the robot in the direction of the robot's movement. The smaller this value, the smaller the distance between the two in the direction of the robot's movement, and the greater the forward pressure exerted by the obstacle on the robot.

[0040] Furthermore, based on the lateral distance and lateral approach velocity, the lateral pressure sub-item of the obstacle on the robot is obtained. Specifically, the quantification formula for the lateral pressure sub-item of the obstacle on the robot is as follows: In the formula, This indicates the degree of lateral pressure exerted on the robot by the obstacle; it is the lateral pressure sub-item. Indicates the speed of the obstacle; Represents the sine function; Indicates the angle between the direction of obstacle movement and the direction of robot movement; Indicates the straight-line distance between the obstacle and the robot; This represents a linear normalization function, such as a max-min normalization function, used to normalize lateral compression values ​​to... Within the range, the maximum and minimum values ​​in the maximum-minimum normalization function are determined based on the extreme values ​​of the lateral compression values ​​in the historical monitoring database. When the normalized object is greater than the maximum value, the normalization result is limited to 1, and when the normalized object is less than the minimum value, the normalization result is limited to 0. 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.

[0041] This represents the lateral approach speed of the obstacle and the robot in the robot's direction of travel. The larger the value, the faster the obstacle and the robot are expected to approach each other, and the greater the lateral pressure exerted by the obstacle on the robot. This represents the lateral distance between the obstacle and the robot in the robot's direction of travel. The smaller this value, the smaller the distance between the obstacle and the robot's expected passage, and the greater the lateral pressure exerted by the obstacle on the robot.

[0042] Furthermore, based on the obstacle's occupancy data of the docking area at the target table location, and combined with the maximum permissible occupancy of the docking area at the target table location, the obstacle's docking pressure sub-item for the robot is obtained. Specifically, the quantification formula for the obstacle's docking pressure sub-item for the robot is as follows: In the formula, This indicates the degree of pressure exerted by obstacles on the robot's docking, i.e., the docking pressure sub-item; This indicates the maximum allowable space occupied in the docking area next to the target table. This indicates the current occupancy of the docking area next to the target table. This represents a linear normalization function, such as a max-min normalization function, used to normalize docking pressure values ​​to a normal value. Within the range, the maximum and minimum values ​​in the maximum-minimum normalization function are determined based on the extreme values ​​of the parking pressure values ​​in the historical monitoring database. When the normalized object is greater than the maximum value, the normalization result is limited to 1, and when the normalized object is less than the minimum value, the normalization result is limited to 0.

[0043] The closer the current occupancy of the docking area at the target table location is to the maximum allowed occupancy, the smaller the available docking space at the target table location, and the greater the pressure exerted by the obstacle on the robot's docking.

[0044] Finally, based on the forward pressure sub-item, lateral pressure sub-item, and docking pressure sub-item, and by introducing intrusion type weights, the degree of intrusion pressure exerted by the obstacle on the robot is obtained. Specifically, the formula for calculating the degree of intrusion pressure exerted by the obstacle on the robot is: In the formula, This indicates the degree of intrusion and pressure exerted by the obstacle on the robot; This indicates the degree of forward pressure exerted by the obstacle on the robot; Indicates the degree of lateral pressure exerted on the robot by the obstacle; This indicates the degree of pressure exerted by obstacles on the robot's docking. , , Both represent intrusion type weights, and .

[0045] Based on the robot's current food delivery stage , , The value needs to be set. If the robot is currently in the passage or departure phase, its need for continuous forward space is high, but its need for docking space at the table is low, therefore the value needs to be increased. Value and reduction Value (e.g.) , If the robot is currently in the docking phase, its demand for docking space at the table is high, but its demand for continuous forward space is low; therefore, it needs to increase... Value and reduction Value (e.g.) , If the robot is currently in the food-fetching phase, it has a higher need for lateral clearance, but a lower need for continuous forward space and tableside parking space. Therefore, it needs to improve... Value (e.g.) , , ).

[0046] When multiple obstacles intrude simultaneously, the total degree of intrusion pressure exerted on the robot by all obstacles is the sum of the degree of intrusion pressure exerted on the robot by each obstacle, denoted as . And determine the overall degree of intrusion and oppression. If the value is greater than 2, then perform pre-deceleration (lower the maximum allowable linear velocity and maximum allowable acceleration limit of the local path planning algorithm) or a brief stop.

[0047] By quantifying the oppression mechanisms of different intrusion behaviors through sub-items, a weighted comprehensive impact assessment of complex intrusions is achieved, avoiding the one-sidedness of a single judgment.

[0048] It should be noted that before calculating the degree of intrusion and pressure, the direction of motion of the obstacle relative to the robot in the robot's direction of travel is first obtained, that is, the longitudinal approach velocity of the obstacle and the robot in the robot's direction of travel is calculated. The relationship with 0 is used to determine whether the obstacle is within the tableside stopping area at the target table location; if the obstacle and the robot are stationary or moving away from each other in the robot's direction of travel, then... Furthermore, since the obstacle is not within the table-edge resting area at the target table location, the degree of intrusion and pressure is directly determined. The calculation of the intrusion and oppression level formula is skipped. Additionally, when an obstacle exceeds the radar detection radius, the intrusion and oppression level is also directly determined. Skip the calculation of the degree of intrusion and compression.

[0049] S500: Detects the continuous barrier-free distance in the tangent direction of the current planned trajectory and the barrier-free space span on both sides in the normal direction, as well as the barrier-free space area of ​​the table-side resting area, to determine the current level of spatial support.

[0050] Different scenarios have different spatial requirements. For example, the main passageway needs continuous forward space, the tableside area needs parking space, and the corner passing area needs space to avoid obstacles on both sides. Therefore, a single distance cannot cover all scenario requirements. It is necessary to consider the compression performance of the overall available space in order to make obstacle avoidance decisions in advance, thereby improving the passability of subsequent plans.

[0051] If the available forward length continues to shorten, it indicates that the continuous forward space is being compressed; if the available width on both sides decreases significantly, it indicates that the lateral avoidance space is being compressed; if the reachable area of ​​the table edge is occupied, it indicates that the docking space is being compressed; therefore, it is necessary to analyze the available space's ability to support the robot's needs.

[0052] Based on the above analysis, in the embodiments of the present invention, the current level of spatial support is determined by detecting the continuous barrier-free distance in the tangent direction of the current planned trajectory and the barrier-free space span on both sides in the normal direction, as well as detecting the area of ​​the barrier-free space region of the table-side resting area. Further aspects include: First, detect the continuous barrier-free distance in the tangent direction of the current planned trajectory and the barrier-free space span on both sides in the normal direction, as well as the area of ​​the barrier-free space in the table-side resting area.

[0053] Then, based on the continuous barrier-free distance in the tangential direction and the continuous barrier-free standard requirement distance in the tangential direction, the forward space adequacy is obtained; at the same time, based on the barrier-free space span on both sides of the normal direction and the barrier-free standard requirement space span on both sides of the normal direction, the lateral space adequacy is obtained; and furthermore, based on the barrier-free space area of ​​the table-side resting area and the barrier-free standard requirement space area of ​​the table-side resting area, the table-side resting space adequacy is obtained. Finally, based on the adequacy of forward space, lateral space, and tableside resting space, and by introducing scene weighting coefficients, the current level of spatial support is determined. Specifically, the formula for calculating the current level of spatial support is: In the formula, Indicates the current level of spatial support; This represents the continuous unobstructed distance along the tangent direction of the current planned trajectory; This indicates the unobstructed space span to the left of the current planned trajectory normal direction; This indicates the unobstructed space span to the right of the current planned trajectory normal direction; This indicates the area of ​​the accessible space in the tableside rest area; This indicates the standard distance required for continuous accessibility in the forward space; This indicates the spatial span required for accessibility standards on both sides; This indicates the area of ​​the accessible standard requirements for the tableside rest area; Indicates the forward weighting coefficient. Indicates the lateral weighting coefficient. This represents the docking weighting coefficient, and tanh represents the tanh normalization function.

[0054] Based on the robot's current food delivery stage , , The value needs to be set. If the robot is currently in the passage or departure phase, its need for continuous forward space is high, but its need for docking space at the table is low, therefore the value needs to be increased. Value and reduction Value (e.g.) , If the robot is currently in the docking phase, its demand for docking space at the table is high, but its demand for continuous forward space is low; therefore, it needs to increase... Value and reduction Value (e.g.) , If the robot is currently in the food-fetching phase, it has a higher need for lateral clearance, but a lower need for continuous forward space and tableside parking space. Therefore, it needs to improve... Value (e.g.) , , Additionally, for special handling, if a meeting scenario is detected during the passage phase (meeting at a corner, in the main passage area, or in the return passage area), the lateral weight will be adjusted based on the weighting coefficients set for each stage of the passage phase. Increase by 0.2, forward weights Reduce by 0.2 to match the lateral space requirements when passing other vehicles.

[0055] Based on the robot's current food delivery stage settings , , If the robot is currently in the passage or departure phase, the requirement for continuous forward space is relatively high, therefore, it is set... , , If the robot is currently in the docking phase, its need for docking space on the table is relatively high, therefore it is set... , , If the robot is currently in the food-fetching phase, it requires a higher level of space to maneuver on both sides, therefore it is set... , , Additionally, for special handling, if oncoming traffic is detected during the passage phase (either in a corner passing area, or in the main passage area or return passage area), then a special setting is configured. , .

[0056] When quantifying the level of spatial support, it is necessary to match the key needs of the robot's current scenario to achieve precise alignment between real-time spatial supply and robot access needs, avoiding the one-sidedness of measuring from a single dimension.

[0057] S600: Based on the degree of spatial support, the degree of stable passage requirement, and the degree of intrusion and pressure, obtain the robot's current safety status maintenance value and output the robot's current obstacle avoidance control strategy.

[0058] Based on the degree of spatial support, stable passage requirements, and intrusion pressure, the robot's current safety status maintenance value is obtained, and the robot's current obstacle avoidance control strategy is output. Further details include: First, the safety status is jointly determined by demand, pressure, and support. Therefore, based on the degree of spatial support, the demand for stable passage, and the degree of intrusion pressure, the robot's current safety status maintenance value is obtained. This value comprehensively reflects the current safety level. Specifically, the formula for calculating the robot's current safety status maintenance value is: In the formula, This indicates the robot's current safety status maintenance value; Indicates the current level of spatial support (values ​​range from 0 to 1); This indicates the robot's current stable passage demand (values ​​range from 0 to 1). This represents the total degree of intrusion and pressure exerted on the robot by all obstacles (with a value ranging from 0 to 2).

[0059] The value range is from -2 to 1. The larger the value, the better the safety status.

[0060] Then, based on the robot's current safety status maintenance value, the robot's current obstacle avoidance control strategy is output. The specific implementation method is as follows: A first safety state threshold and a second safety state threshold are preset, with the first safety state threshold being greater than the second safety state threshold. The first and second safety state thresholds can be set as needed. Specifically, based on the above calculation method for the safety state maintenance value, this embodiment sets the first safety state threshold to 0.6 and the second safety state threshold to 0.3.

[0061] If the safety state maintenance value is greater than or equal to the first safety state threshold, that is... If so, the robot is currently in a safe state and is proceeding normally.

[0062] If the safety state maintenance value is less than the first safety state threshold and greater than or equal to the second safety state threshold, that is... If the robot is in a warning state, then an obstacle avoidance strategy is formulated based on the adequacy of forward space, lateral space, and tableside docking space, as well as the performance of the forward pressure sub-item, lateral pressure sub-item, and docking pressure sub-item. More specifically, the obstacle avoidance strategy is as follows: If the obstacle is characterized by insufficient forward continuous space (forward space sufficiency is less than the preset forward space sufficiency threshold of 0.6), then pre-deceleration (lowering the maximum allowable linear velocity and maximum allowable acceleration upper limit of the local path planning algorithm) or brief stopping is performed; if the obstacle is characterized by pressure on one side while the other side still has support margin (lateral pressure degree is greater than the preset lateral pressure degree threshold of 0.6, and lateral space sufficiency is greater than the preset lateral space sufficiency threshold of 0.5), then lateral offset detour is performed (increasing the cost inflation radius in the local planner and increasing the lateral sampling step size parameter); if the obstacle is characterized by insufficient space in the table-side parking area (table-side parking space sufficiency is less than the table-side parking space sufficiency threshold of 0.6), then switching to a nearby reachable table-side parking position; if the overall space support is insufficient (space support degree... (Less than the spatial support threshold of 0.6) and persistent intrusive pressure (intrusive pressure level) If the intrusion pressure level exceeds the threshold of 0.3 and lasts for more than 5 seconds, then either wait for concession or perform a global replanning.

[0063] If the safety state maintenance value is less than the second safety state threshold, that is If the robot is in a dangerous state, it will execute an emergency stop strategy.

[0064] Special handling is required in the following situations: if the ranging sensor detects that the distance to an obstacle is less than the preset absolute safety radius (e.g., 0.2m), the chassis hardware will be triggered to stop immediately, even if the robot is in a warning state, and the smooth obstacle avoidance strategy will not be executed; or, if the food being carried is a highly sensitive type (e.g., soup), even if it is in a safe state, if the forward space fluctuates by ≥20% (the forward space adequacy changes by more than 20% per unit time), pre-deceleration will be triggered to prevent the food from shaking.

[0065] This embodiment also provides an active obstacle avoidance control system for a catering service robot, including: a memory and a processor, wherein: Memory, used to store program code; The processor reads the program code stored in the memory and executes it to determine the current food delivery stage of the robot and obtain the basic spatial requirement characteristics of the current food delivery stage; it collects the robot's three-axis acceleration data, analyzes the robot's current sway amplitude and recovery time, and obtains the robot's current sway sensitivity coefficient; it combines the basic spatial requirement characteristics and the sway sensitivity coefficient to obtain the robot's current stable passage requirement; it detects the position, speed, and occupancy data of obstacles, and analyzes the forward pressure, lateral pressure, and docking pressure sub-items of obstacles on the robot to obtain the degree of intrusion pressure of obstacles on the robot; it detects the continuous unobstructed distance in the tangent direction of the current planned trajectory and the unobstructed space span on both sides in the normal direction, as well as the unobstructed space area of ​​the table-side docking area, to determine the current spatial support level; based on the spatial support level, stable passage requirement, and intrusion pressure level, it obtains the robot's current safety state maintenance value and outputs the robot's current obstacle avoidance control strategy.

[0066] Furthermore, the processor includes a stable passage demand analysis module, an intrusion pressure analysis module, a space support analysis module, and an obstacle avoidance control module, wherein: The stable passage demand analysis module is used to determine the current food delivery stage of the robot and obtain the basic space demand characteristic value of the current food delivery stage; and to collect the robot's three-axis acceleration data, analyze the robot's current sway amplitude and recovery time, and obtain the robot's current sway sensitivity coefficient; and then, by combining the basic space demand characteristic value and the sway sensitivity coefficient, obtain the robot's current stable passage demand degree. The intrusion and pressure analysis module is used to detect the position, speed, and occupancy data of obstacles, and analyze the forward pressure, lateral pressure, and docking pressure of obstacles on the robot to obtain the degree of intrusion and pressure of obstacles on the robot. The spatial support analysis module is used to detect the continuous barrier-free distance in the tangent direction of the current planned trajectory and the barrier-free space span on both sides in the normal direction, as well as the area of ​​the barrier-free space in the table-side resting area, to determine the current spatial support level. The obstacle avoidance control module is used to obtain the robot's current safety status maintenance value based on the degree of spatial support, the degree of stable passage requirement, and the degree of intrusion and pressure, and output the robot's current obstacle avoidance control strategy.

Claims

1. A method for active obstacle avoidance control of a catering service robot, characterized in that, The method includes: Determine the current food delivery stage of the robot and obtain the basic spatial requirement characteristics of the current food delivery stage; Collect the robot's three-axis acceleration data, analyze the robot's current sway amplitude and recovery time, and obtain the robot's current sway sensitivity coefficient; By combining the basic spatial demand characteristic value and the sway sensitivity coefficient, the current stable passage demand of the robot is obtained; The system detects the position, speed, and occupancy data of obstacles, and analyzes the forward pressure, lateral pressure, and docking pressure of obstacles on the robot to obtain the degree of intrusion and pressure of obstacles on the robot. The system detects the continuous barrier-free distance along the tangent direction of the current planned trajectory and the barrier-free space span on both sides along the normal direction, as well as the barrier-free space area of ​​the table-side resting area, to determine the current level of spatial support. Based on the spatial support level, the stable passage requirement level, and the intrusion pressure level, the robot's current safety status maintenance value is obtained, and the robot's current obstacle avoidance control strategy is output.

2. The active obstacle avoidance control method for a catering service robot according to claim 1, characterized in that, Determining the current stage of the robot's food delivery process includes: Divide the restaurant into different scene areas on a static map; After receiving the food delivery task, the robot obtains the location of the target table and the robot's current location. It then matches the current location with the restaurant's static map to determine the scene area where the robot is currently located. Combining the task progress and the target table location, it determines the current stage of the robot's food delivery.

3. The active obstacle avoidance control method for a catering service robot according to claim 2, characterized in that, The food delivery phase includes the food pickup phase, the departure phase, the passage phase, and the stop phase; the scene areas include the food dispensing area, the main passage area, the corner passing area, the table-side stopping area, and the return passage area. The robot's current location is then matched with the restaurant's static map to determine its current scene area. Combined with the task progress and the target table's location, the robot's current food delivery stage is determined, including: If the robot is currently in the food dispensing area and has not yet picked up any food, then the robot is currently in the food picking phase. If the robot is currently in the food dispensing area and has already picked up a meal, then the robot is considered to be in the departure phase. If the robot is currently located in the main passage area, corner meeting area or return passage area, and is moving towards the target table or base station, then the robot is determined to be in the passage phase. If the robot's current position is in the tableside docking area and the distance to the target table position is less than the preset distance, then the robot is determined to be in the docking phase.

4. The active obstacle avoidance control method for a catering service robot according to claim 1, characterized in that, Collect the robot's three-axis acceleration data, analyze the robot's current sway amplitude and recovery time, and obtain the robot's current sway sensitivity coefficient, including: Collect the robot's three-axis acceleration data, calculate the root mean square of the three-axis acceleration, and obtain the robot's current sway amplitude; The time required for the sway amplitude to decrease to below 10% of the initial maximum sway amplitude is monitored to obtain the robot's current stabilization time. Based on the sway amplitude and the stabilization time, combined with the maximum permissible sway amplitude and maximum permissible stabilization time at the factory calibration, and by introducing sway weight and stabilization weight, the robot's current sway sensitivity coefficient is obtained.

5. The active obstacle avoidance control method for a catering service robot according to claim 1, characterized in that, By combining the aforementioned basic spatial demand characteristic value and the sway sensitivity coefficient, the robot's current stable passage demand is obtained, including: Based on the basic space demand characteristic value and the sway sensitivity coefficient, combined with the maximum basic space demand characteristic value and the maximum sway sensitivity coefficient, and by introducing space demand weight and sway sensitivity weight, the current stable passage demand of the robot is obtained.

6. The active obstacle avoidance control method for a catering service robot according to claim 3, characterized in that, The system detects the position, velocity, and occupancy data of obstacles, and analyzes the forward pressure, lateral pressure, and docking pressure sub-items exerted by the obstacles on the robot to obtain the degree of intrusion and pressure on the robot, including: The system detects the position and speed of obstacles and their occupancy data in the docking area at the target table location. Combined with the robot's current position and running speed, it obtains the longitudinal distance, longitudinal approach speed, lateral distance, and lateral approach speed between the obstacle and the robot in the robot's direction of travel. Based on the longitudinal distance and the longitudinal approach speed, the forward pressure sub-item of the obstacle on the robot is obtained; Based on the lateral distance and the lateral approach speed, the lateral pressure sub-item of the obstacle on the robot is obtained; Based on the data on the occupancy of the obstacle in the docking area at the target table location, and combined with the maximum allowable occupancy of the docking area at the target table location, the sub-item of the obstacle's docking pressure on the robot is obtained; Based on the forward pressure sub-item, the lateral pressure sub-item, and the docking pressure sub-item, and by introducing an intrusion type weight, the degree of intrusion pressure exerted by the obstacle on the robot is obtained.

7. The active obstacle avoidance control method for a catering service robot according to claim 1, characterized in that, The system detects the continuous barrier-free distance along the tangent direction of the current planned trajectory and the barrier-free space span on both sides along the normal direction, as well as the area of ​​the barrier-free space in the table-side resting area, to determine the current level of spatial support, including: The system detects the continuous barrier-free distance along the tangent of the current planned trajectory and the barrier-free space span on both sides along the normal direction, as well as the area of ​​the barrier-free space in the tableside resting area. The forward space adequacy is obtained by combining the continuous unobstructed distance in the tangential direction with the standard unobstructed distance requirement in the tangential direction. Based on the unobstructed space span on both sides of the normal direction, and combined with the unobstructed standard requirement space span on both sides of the normal direction, the lateral space adequacy is obtained. Based on the area of ​​the accessible space in the tableside parking area, and combined with the area of ​​the accessible standard requirements for the tableside parking area, the sufficiency of the tableside parking space is obtained. Based on the aforementioned forward space adequacy, lateral space adequacy, and tableside docking space adequacy, and by introducing scene weighting coefficients, the current level of spatial support is determined.

8. The active obstacle avoidance control method for a catering service robot according to claim 7, characterized in that, Based on the spatial support level, the stable passage requirement, and the intrusion pressure level, the robot's current safety status maintenance value is obtained, and the robot's current obstacle avoidance control strategy is output, including: Based on the spatial support level, the stable passage requirement level, and the intrusion pressure level, the current safety status maintenance value of the robot is obtained; A first security state threshold and a second security state threshold are preset; the first security state threshold is greater than the second security state threshold. If the safety status maintenance value is greater than or equal to the first safety status threshold, the robot is currently in a safe state and is proceeding normally. If the safety status maintenance value is less than the first safety status threshold and greater than or equal to the second safety status threshold, the robot is currently in a warning state. Then, based on the forward space adequacy, the lateral space adequacy, the tableside docking space adequacy, and the performance of the forward pressure sub-item, the lateral pressure sub-item, and the docking pressure sub-item, an obstacle avoidance strategy is formulated. If the safety status maintenance value is less than the second safety status threshold, the robot is currently in a dangerous state and an emergency stop strategy is executed.

9. An active obstacle avoidance control system for a catering service robot, characterized in that, The system includes: a memory and a processor, wherein: The memory is used to store program code; The processor is configured to read program code stored in the memory and execute the method as described in any one of claims 1 to 8.

10. The active obstacle avoidance control system for the catering service robot according to claim 9, characterized in that, The processor includes: The stable passage demand analysis module is used to determine the current food delivery stage of the robot and obtain the basic space demand characteristic value of the current food delivery stage; and to collect the robot's three-axis acceleration data, analyze the robot's current swaying amplitude and recovery time, and obtain the robot's current swaying sensitivity coefficient; and then, by combining the basic space demand characteristic value and the swaying sensitivity coefficient, obtain the robot's current stable passage demand degree. The intrusion and pressure analysis module is used to detect the position, speed, and occupancy data of obstacles, and analyze the forward pressure, lateral pressure, and docking pressure of obstacles on the robot to obtain the degree of intrusion and pressure of obstacles on the robot. The spatial support analysis module is used to detect the continuous barrier-free distance in the tangent direction of the current planned trajectory and the barrier-free space span on both sides in the normal direction, as well as the area of ​​the barrier-free space in the table-side resting area, to determine the current spatial support level. The obstacle avoidance control module is used to obtain the robot's current safety status maintenance value based on the spatial support level, the stable passage requirement level, and the intrusion pressure level, and output the robot's current obstacle avoidance control strategy.

Images