Intelligent navigation method for robot

Abstract

The invention relates to the technical field of robot navigation, in particular to an intelligent navigation method for a robot, and solves the problem of fuzzy recognition of irregular obstacles in a traditional method by acquiring an obstacle image in a moving track of the robot in real time, preprocessing the image, extracting motion features and classifying motion types. According to the method, differential judgment logic is adopted for the obstacles of different motion types, the passing mode of the regularly-moving obstacles is determined by integrating the relative motion distance and the detour difficulty, excessive conservative is avoided, and on the other hand, the passing mode of the irregularly-moving obstacles is determined by integrating the relative motion distance and the detour difficulty. An operation uncertainty characterization value is introduced to quantify the motion disorder degree, and by combining real-time relative distance judgment, the detour safety is ensured, so that efficiency reduction caused by excessive avoidance of a low-risk obstacle is prevented, and collision caused by forced detour of a high-risk obstacle is also avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of robot navigation, and in particular to an intelligent navigation method for a robot. Background Art

[0002] In the field of autonomous navigation for robots (such as service robots and delivery robots), avoiding dynamic obstacles (especially those with irregular movements, such as pedestrians and pets) is a core technical challenge. Most existing algorithms are designed for static obstacles (such as fixed facilities) or regularly moving obstacles (such as vehicles traveling along a fixed path), achieving detours through preset safe distances or trajectory prediction. However, for irregularly moving obstacles (such as pedestrians who suddenly change direction or pets running randomly), whose movement direction and speed frequently change suddenly, traditional methods struggle to accurately predict their trajectories, which can easily lead to collisions or overly conservative detours (such as frequent emergency stops, which reduce traffic efficiency).

[0003] Chinese patent application number: CN202411690612.8 discloses an intelligent navigation path planning system for a robot, including a controller, a throwing module, a driving wheel, a radio frequency identification module and a positioning module; the input end of the controller is connected to a laser radar, a camera, a ranging sensor, a cliff sensor and a collision sensor, which are used to realize the shooting and identification of the environment, obstacles and height differences around the travel route, and the output end of the controller is connected to a driving wheel. When there is an obstacle and the route needs to be replanned, the present invention throws the RFID tag onto the ground. During the subsequent inspection process, as long as the equipment identification module recognizes that the RFID tag still exists, it is determined that the obstacle still exists. There is no need to replan the route again and the backup inspection route is directly used. When the obstacle and the RFID tag are cleared together and the tag cannot be identified, the original planned route is used, reducing the number of planned routes and improving inspection efficiency.

[0004] However, the prior art still has the following problems: For irregular moving obstacles, their movement direction and speed frequently change suddenly, and traditional methods are difficult to accurately predict their trajectory, which can easily lead to collisions or overly conservative detours. Summary of the Invention

[0005] To this end, the present invention provides an intelligent navigation method for a robot to overcome the problem in the prior art that for irregular moving obstacles, the movement direction and speed frequently change suddenly, and traditional methods are difficult to accurately predict the trajectory, which easily leads to collisions or overly conservative detours.

[0006] To achieve the above objectives, the present invention provides an intelligent navigation method for a robot. The method comprises: Step S1, obtaining an obstacle image in the robot's running trajectory in real time, preprocessing the obtained obstacle image, extracting the obstacle's motion characteristics, and determining the obstacle's motion type based on the obstacle's motion characteristics; Step S2, determining the robot's obstacle avoidance method based on the determined movement type of the obstacle, includes: Determine the robot's passage method based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle; Alternatively, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the trajectory of the irregularly moving obstacle and the obstacle's operational uncertainty characterization value; Step S3, when the passage mode is determined to be bypassing, analyzing the movement trajectory of the irregularly moving obstacle to determine the range of the movement trajectory, and selecting the bypass trajectory with the shortest bypass distance and the least difficulty based on the robot position; Step S4: determining the travel trajectory according to the determined travel mode.

[0007] Furthermore, in step S1, determining the movement type of the obstacle based on the movement characteristics of the obstacle includes: Based on the pre-processed obstacle image, the motion feature data of the obstacle at several consecutive time nodes are extracted according to a predetermined interval to obtain a time series feature sequence; Analyzing the motion feature data in the time series feature sequence and calculating core feature data; Compare the core feature data with the preset thresholds respectively; Based on the comparison result, it is determined whether there is irregular core feature data, and when irregular core feature data exists, the movement type of the obstacle is determined to be irregular movement.

[0008] Furthermore, the core feature data includes speed standard deviation, trajectory motion deviation and direction change frequency.

[0009] Furthermore, in step S2, determining the robot's obstacle avoidance method based on the determined movement type of the obstacle includes: If the obstacle's motion type is regular, the robot's passage method is determined based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle. If the movement type of the obstacle is irregular movement, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the movement trajectory of the irregularly moving obstacle and the obstacle's movement uncertainty characterization value.

[0010] Furthermore, in step S2, the difficulty of the robot circumventing the obstacle is determined as follows: Determine the maximum change in the actual traversable space when the robot circumvents the obstacle, Determine the angle between the obstacle's running direction and the robot's current running direction, and obtain the curvature of the detour path. Calculate the ratio of the maximum change in the actual traversable space to the preset maximum change to obtain the first sub-detour difficulty parameter. Calculate the ratio of the detour path curvature to the preset detour path curvature to obtain the second sub-detour difficulty parameter, The first bypass difficulty parameter and the second sub-bypass difficulty parameter are weighted and summed to obtain the bypass difficulty.

[0011] Furthermore, in step S2, the robot's passage mode is determined based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle, including: If the relative motion distance is greater than or equal to the preset relative motion distance, it is determined that the robot is traveling along the original driving path; If the relative motion distance is less than the preset relative motion distance and the detour difficulty is less than or equal to the preset detour difficulty, it is determined that the robot has detoured; If the relative motion distance is smaller than the preset relative motion distance and the detour difficulty is greater than the preset detour difficulty, it is determined that the robot is waiting to pass.

[0012] Furthermore, in step S2, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the running trajectory of the irregular motion obstacle and the obstacle's running uncertainty characterization value, including: If the real-time relative distance is greater than the preset real-time relative distance, it is determined that the robot is traveling along the original driving path; If the real-time relative distance is less than the preset real-time relative distance and the obstacle's operational uncertainty representation value is less than or equal to the preset operational uncertainty representation value, then it is determined that the robot is bypassing the obstacle; If the real-time relative distance is smaller than the preset real-time relative distance and the obstacle's operational uncertainty characterization value is larger than the preset operational uncertainty characterization value, it is determined that the robot is waiting to pass.

[0013] Furthermore, the determination of the operational uncertainty characterization value of the obstacle includes: Calculate the ratio of the obstacle's direction change rate to a preset direction change rate to obtain a first operational uncertainty sub-parameter. Calculate the ratio of the speed change rate of the obstacle to the preset speed change rate to obtain the second operation uncertainty sub-parameter, The first operational uncertainty sub-parameter and the second operational uncertainty sub-parameter are weightedly summed to obtain an operational uncertainty characterization value.

[0014] Furthermore, the determination of the real-time relative distance includes: Collect robot position data and obstacle position data in real time, and calculate the spatial straight-line distance as the initial real-time relative distance based on the robot position data and the obstacle position data; Obtain obstacle movement direction and speed data, and calculate the relative distance change per unit time; The initial real-time relative distance is corrected based on the relative distance change to obtain the real-time relative distance.

[0015] Furthermore, in step S3, determining the range of the motion trajectory includes: Based on the image data preprocessed in step S1, the position coordinate sequence of the obstacle is extracted to form a discrete motion trajectory point set; Obtain the maximum and minimum displacements of each coordinate axis to obtain the boundary points; Use smooth curves to connect each boundary point to obtain the range of motion trajectory; The detour trajectories with the shortest distance and the least difficulty based on the robot's position include: Determine the starting point and end point based on the robot's real-time position coordinates and the target coordinates; Generate several candidate detour trajectories that meet the constraints; The detour difficulties of the candidate trajectories are evaluated, and the candidate trajectory with the minimum detour difficulty is selected as the detour trajectory.

[0016] Compared with the existing technology, the present invention has the beneficial effect of solving the problem of "ambiguous identification of irregular obstacles" in traditional methods by preprocessing images, extracting motion features (such as the rate of change of direction and speed), and classifying motion types (regular / irregular motion). This avoids erroneous decisions caused by misjudging obstacle types. The present invention adopts differentiated judgment logic for different types of obstacles: for regularly moving obstacles, the method of passage is determined by comprehensively considering relative motion distance and detour difficulty (for example, if the distance is sufficient and the detour difficulty is low, the method is directly detoured; otherwise, it waits), thus avoiding excessive conservatism. On the other hand, for irregularly moving obstacles, an "operational uncertainty characterization value" is introduced to quantify the degree of motion chaos. Combined with real-time relative distance judgment (for example, if the uncertainty value is below a preset threshold and the distance is sufficient, the method is detoured; otherwise, it waits), this ensures detour safety. This prevents both excessive avoidance of low-risk obstacles (such as slowly walking pedestrians), which may lead to reduced efficiency, and forced detour of high-risk obstacles (such as rapidly changing pets), which may cause collisions.

[0017] Furthermore, in the present invention, by "extracting feature data of continuous time nodes at equal intervals" (such as extracting once every 0.5 seconds), a time series feature sequence is formed, rather than relying on single-frame or short-time data. This can capture the "dynamic trend" of obstacle movement (such as pedestrians gradually changing from regular walking to random pauses), avoid misjudgment due to instantaneous data (such as a short pause in regular movement being directly classified as irregular), and is particularly suitable for scenarios with gradual changes in motion state (such as a robot that slowly changes direction).

[0018] Furthermore, the present invention considers that relative motion spacing can directly reflect collision risk, and that the difficulty of bypassing (combined with changes in traversable space and path curvature) ensures the feasibility of bypassing (direct bypassing when the difficulty is low, avoiding unnecessary waiting). Therefore, regularly moving obstacles are analyzed based on a comprehensive consideration of relative motion spacing and bypassing difficulty, thereby improving control accuracy for regularly moving obstacles, avoiding overly conservative treatment of regularly moving obstacles, and improving passage efficiency while ensuring safety. On the other hand, considering that real-time relative spacing ensures a physical safety buffer when the robot reaches the obstacle trajectory (it can go straight without additional operation when the spacing is sufficient), the operational uncertainty characterization value is determined by the direction change rate and speed change rate, and then the "chaos" of the obstacle movement is measured (detoxification is possible when the uncertainty value is low, and waiting is required when it is high to avoid collision due to prediction failure). This ensures that the potential mutation risk of the obstacle is not ignored due to temporary stability, nor is frequent emergency stops caused by excessive fear of uncertainty (for example, there is no need to wait for slightly shaking obstacles), thus balancing safety and smooth passage.

[0019] Furthermore, the present invention converts the fluctuation degree of space width into a calculable value through the "ratio of the preset maximum change amount to the actual change amount", and quantifies the degree of path curvature through the "ratio of the actual detour curvature to the preset curvature", thereby avoiding the subjectivity of the traditional "judging the difficulty by experience", making the detour difficulty assessment standardized and reproducible, and providing a unified basis for subsequent "straight / detour / wait" decisions. By assigning weights to the two sub-parameters (such as a higher weight for spatial changes in narrow scenes and a higher weight for curvature in open scenes), a comprehensive assessment of the detour difficulty is made, focusing on both spatial stability (avoiding "being able to pass but the space is wide or narrow, resulting in a collision") and path feasibility (avoiding "enough space but the path turns too sharply, resulting in loss of control"), achieving balanced control of multi-dimensional risks and reducing the collision risk during the detour process.

[0020] Furthermore, in the present invention, when the relative motion distance is greater than a preset value, it is directly determined that "the original path is passable" without additional operation. When the distance is insufficient, the bypass difficulty is compared with the preset difficulty to distinguish between bypassability and waiting. When the bypass difficulty is relatively low, it means that the space around the obstacle is stable (small change) and the path is flat (low curvature), and the robot can bypass it safely. When the bypass difficulty is relatively high, it means that the space changes drastically or the path turns too sharply, and forced bypassing may cause a collision. By looking at the hierarchical logic of first looking at the distance and then the difficulty, the limitations of single indicator judgment are avoided, making the decision more comprehensive and safe.

[0021] Furthermore, in the present invention, for obstacles with sufficient spacing but easy to bypass (such as a tracking robot traveling at a constant speed in the distance), the original path is directly used to pass through, avoiding unnecessary detours or waiting, thereby improving passage efficiency. For obstacles with insufficient spacing but easy to bypass (such as slowly moving shelves with sufficient space on both sides), the shortest path is used to bypass them, which neither affects efficiency nor avoids task delays caused by waiting. For obstacles with insufficient spacing and high difficulty to bypass (such as a robotic arm moving back and forth at high speed and with a narrow space around it), the choice is to wait to pass through, ensuring safety and avoiding equipment damage or collision due to forced detours. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Flowchart of the intelligent navigation method for a robot according to the present invention; Figure 2 A flow chart for determining the robot's obstacle avoidance method; Figure 3 A flow chart for determining the passage mode of a robot that moves regularly; Figure 4 A flow chart for determining the passage mode of a robot that moves erratically. DETAILED DESCRIPTION

[0023] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0024] It should be noted that the data in this embodiment are obtained by comprehensive analysis and evaluation of the historical data of the six months before the current determination and the corresponding historical determination results by the system of the present invention. It can be understood by those skilled in the art that the system of the present invention can determine the above parameters for each of the above parameters by selecting the value with the highest proportion as the preset standard parameter based on the data distribution, using weighted summation to use the obtained value as the preset standard parameter, substituting each historical data into a specific formula and using the value obtained by the formula as the preset standard parameter, or other selection methods, as long as the system of the present invention can clearly define the different specific situations in the single determination process through the obtained values.

[0025] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0026] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside", and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0027] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0028] See also Figure 1 As shown, it is a flow chart of the intelligent navigation method for robots of the present invention.

[0029] The intelligent navigation method for a robot provided in this embodiment includes: Step S1, obtaining an obstacle image in the robot's running trajectory in real time, preprocessing the obtained obstacle image, extracting the obstacle's motion characteristics, and determining the obstacle's motion type based on the obstacle's motion characteristics; Step S2, determining the robot's obstacle avoidance method based on the determined movement type of the obstacle, includes: The robot's passage mode is determined based on the relative motion distance between the obstacle and the robot and the difficulty of the robot to bypass the obstacle. Alternatively, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the trajectory of the irregularly moving obstacle and the obstacle's operational uncertainty characterization value; Step S3, when the passage mode is determined to be bypassing, analyzing the movement trajectory of the irregularly moving obstacle to determine the range of the movement trajectory, and selecting the bypass trajectory with the shortest bypass distance and the least difficulty based on the robot position; Step S4: determining the travel trajectory according to the determined travel mode.

[0030] Specifically, in this embodiment, the position data of the obstacle in the obstacle image is obtained by using a laser radar or a depth camera.

[0031] The present invention solves the problem of "ambiguous identification of irregular obstacles" in traditional methods by preprocessing images, extracting motion features (such as the rate of change of direction and speed), and classifying motion types (regular / irregular motion). This avoids erroneous decisions caused by misjudging the obstacle type. The present invention adopts differentiated judgment logic for obstacles with different motion types: for regularly moving obstacles, the passage method is determined by comprehensively considering the relative motion distance and the difficulty of bypassing (for example, if the distance is sufficient and the bypass difficulty is low, directly bypass it; otherwise, wait), avoiding excessive conservatism. On the other hand, for irregularly moving obstacles, an "operational uncertainty characterization value" is introduced to quantify the degree of motion chaos. Combined with real-time relative distance judgment (for example, if the uncertainty value is below the preset threshold and the distance is sufficient, bypass it; otherwise, wait), the bypass safety is ensured. This prevents both excessive avoidance of low-risk obstacles (such as slowly walking pedestrians) that leads to reduced efficiency, and forced bypass of high-risk obstacles (such as pets that change direction quickly) that may cause collisions.

[0032] Specifically, in step S1, determining the movement type of the obstacle based on the movement characteristics of the obstacle includes: Based on the pre-processed obstacle image, the motion feature data of the obstacle at several consecutive time nodes are extracted according to a predetermined interval to obtain a time series feature sequence; Analyzing the motion feature data in the time series feature sequence and calculating core feature data; Compare the core feature data with the preset thresholds respectively; Based on the comparison result, it is determined whether there is irregular core feature data, and when irregular core feature data exists, the movement type of the obstacle is determined to be irregular movement, and when irregular core feature data does not exist, the movement type of the obstacle is determined to be regular movement.

[0033] Specifically, the core feature data includes speed standard deviation, trajectory motion deviation, and direction change frequency.

[0034] In the present invention, by "extracting feature data of continuous time nodes at equal intervals" (such as extracting once every 0.5 seconds), a time series feature sequence is formed instead of relying on single-frame or short-time data. This can capture the "dynamic trend" of obstacle movement (such as pedestrians gradually changing from regular walking to random pauses), avoid misjudgment due to instantaneous data (such as a short pause in regular movement being directly classified as irregular), and is particularly suitable for scenarios with gradual changes in motion state (such as a robot that slowly changes direction).

[0035] See also Figure 2 As shown, it is a flow chart for determining the robot's obstacle avoidance method.

[0036] Specifically, in step S2, determining the robot's obstacle avoidance method based on the determined movement type of the obstacle includes: If the obstacle's motion type is regular, the robot's passage method is determined based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle. If the movement type of the obstacle is irregular movement, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the movement trajectory of the irregularly moving obstacle and the obstacle's movement uncertainty characterization value.

[0037] The present invention takes into account that the relative motion distance can directly reflect the collision risk, and the detour difficulty (combined with the change in the traversable space and the path curvature) ensures the feasibility of detour (direct detour when the difficulty is low to avoid unnecessary waiting). Therefore, the regular motion obstacles are analyzed based on the comprehensive relative motion distance and detour difficulty, thereby improving the control accuracy for regular motion obstacles, avoiding overly conservative treatment of regular motion obstacles, and improving the passage efficiency while ensuring safety. On the other hand, considering that the real-time relative distance ensures a physical safety buffer when the robot reaches the obstacle trajectory (it can go straight when the distance is sufficient without additional operation), the operation uncertainty characterization value is determined by the direction change rate and speed change rate, and then the "chaos" of the obstacle movement is measured (detour when the uncertainty value is low, and waiting is required when it is high to avoid collision due to prediction failure). Neither will the potential mutation risk of the obstacle be ignored due to temporary stability, nor will frequent emergency stops be made due to excessive fear of uncertainty (for example, there is no need to wait for slightly shaking obstacles), thus balancing safety and smooth passage.

[0038] Specifically, in step S2, the difficulty of the robot circumventing the obstacle is determined as follows: Determine the maximum change in the actual traversable space when the robot circumvents the obstacle, Determine the angle between the obstacle's running direction and the robot's current running direction, and obtain the curvature of the detour path. Calculate the ratio of the maximum change in the actual traversable space to the preset maximum change to obtain the first sub-detour difficulty parameter. Calculate the ratio of the detour path curvature to the preset detour path curvature to obtain the second sub-detour difficulty parameter, The first bypass difficulty parameter and the second sub-bypass difficulty parameter are weighted and summed to obtain the bypass difficulty.

[0039] Specifically, in this embodiment, the preset maximum change is determined by the following method: obtaining typical operating scenario data of several robots, statistically analyzing the range of changes in the width of the passable space when historical obstacles pass through, and recording the maximum change value within the 95% confidence interval; based on the robot's minimum safe passage width (its own width + safety margin, such as a 0.5-meter robot requires a +0.2-meter safety distance, that is, 0.7 meters), the maximum change value in the historical data is corrected: if the historical maximum change value exceeds the robot's safe passage capacity, the safe passage capacity is used as the upper limit; if it does not exceed, the historical value is retained and a 10%-20% redundancy is added; the corrected value is used as the preset maximum change, stored in the robot control system, and can be dynamically adjusted according to scene switching (such as from indoors to outdoors).

[0040] Specifically, in this embodiment, the preset detour path curvature is determined by: extracting the robot's minimum turning radius Rmin (e.g., if the minimum turning radius of a wheeled robot is 1 meter, then its limit curvature K1=1 / Rmin=1rad / m); based on the scene safety requirements, setting the safety factor α (usually 0.6~0.8, balancing flexibility and safety), then the basic curvature threshold K2=α×K1; if the scene is an open space (such as a warehouse) and obstacles are sparsely distributed, the safety factor can be increased (such as 0.8) to allow a larger curvature (more flexible detour); if the scene is a narrow passage (such as a corridor), the safety factor needs to be reduced (such as 0.6) to limit the curvature (to avoid hitting the wall due to excessive bending of the path); the corrected curvature value is used as the preset detour path curvature.

[0041] In the present invention, the fluctuation degree of space width is converted into a calculable value through the "ratio of the preset maximum change amount to the actual change amount", and the degree of path curvature is quantified through the "ratio of the actual detour curvature to the preset curvature", thereby avoiding the subjectivity of the traditional "judging the difficulty by experience", making the detour difficulty assessment standardized and reproducible, and providing a unified basis for subsequent "straight / detour / wait" decisions. By assigning weights to the two sub-parameters (such as a higher weight for spatial changes in narrow scenes and a higher weight for curvature in open scenes), a comprehensive assessment of the detour difficulty is made, focusing on both spatial stability (avoiding "being able to pass but the space is wide or narrow, resulting in a collision") and path feasibility (avoiding "enough space but the path turns too sharply, resulting in loss of control"), achieving balanced control of multi-dimensional risks and reducing the collision risk during the detour process.

[0042] See also Figure 3 As shown, it is a flow chart for determining the passage mode of a robot with regular movements.

[0043] Specifically, in step S2, the robot's passage mode is determined based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle, including: If the relative motion distance is greater than or equal to the preset relative motion distance, it is determined that the robot is traveling along the original driving path; If the relative motion distance is less than the preset relative motion distance and the detour difficulty is less than or equal to the preset detour difficulty, it is determined that the robot has detoured; If the relative motion distance is smaller than the preset relative motion distance and the detour difficulty is greater than the preset detour difficulty, it is determined that the robot is waiting to pass.

[0044] Specifically, in this embodiment, the preset relative motion distance is determined by the following method: preset relative motion distance = robot's shortest braking distance + obstacle's displacement during the robot's braking time + robot's own safety buffer width + environmental redundancy. For example, the robot's maximum speed is 1.2 m / s, the shortest braking distance (decelerating from 1.2 m / s to stop) is 0.8 m (limited by ground friction and braking system), its own width is 0.6 m, and the safety buffer width is 0.5 m (to avoid edge collision). Assume that the obstacle is a pedestrian moving in a uniform straight line with a speed of 1.0 m / s, and the relative distance to the robot is 0.5 m. The movement direction is perpendicular to the robot's path (i.e., the pedestrian crosses the robot's path), and the robot's braking response time is 0.3s (the delay from obstacle detection to braking). The indoor environment is relatively narrow, and sensors (such as cameras and lidar) have a measurement error of ±0.2m, so the environmental redundancy is 0.3m. During the robot's braking time, the displacement of the obstacle = obstacle speed × braking response time = 1.0m / s × 0.3s = 0.3m. The preset relative motion distance = 0.8m (braking distance) + 0.3m (obstacle displacement) + 0.5m (safety buffer) + 0.3m (environmental redundancy) = 1.9m.

[0045] Specifically, in this embodiment, the preset detour difficulty is determined in the following way: taking an indoor service robot as an example, its operating scenario is a corridor with a width of 2-3 meters, the hardware parameters are a minimum turning radius of 0.5 meters, a maximum turning angle of ±30 degrees, and according to the corridor width, the maximum safe change in the traversable space is set to 0.8 meters (to avoid collisions caused by sudden narrowing of the space); based on the minimum turning radius, the preset detour path curvature is calculated to be 0.3 radians / meter (corresponding to a turning angle of 20 degrees, which is within the safe range of the robot's steering performance); in the indoor corridor scenario, , the change in traversable space has a greater impact on safety, so the weight of the first sub-parameter (spatial change) is 0.6, and the weight of the second sub-parameter (path curvature) is 0.4; when the first sub-parameter (actual change / preset change) ≤ 1, and the second sub-parameter (actual curvature / preset curvature) ≤ 1, the detour difficulty is within the safe range; the preset critical value of the detour difficulty is: (1×0.6) + (1×0.4) = 1.0, that is, when the calculated detour difficulty ≤ 1.0, the robot can detour safely; if it is greater than 1.0, it is out of the safe range and needs to wait.

[0046] In the present invention, when the relative motion distance is greater than a preset value, it is directly determined that "the original path is passable" without additional operation. When the distance is insufficient, the bypass difficulty is compared with the preset difficulty to distinguish between bypassability and waiting. When the bypass difficulty is relatively low, it means that the space around the obstacle is stable (small change) and the path is flat (low curvature), and the robot can bypass it safely. When the bypass difficulty is relatively high, it means that the space changes drastically or the path turns too sharply, and forced bypassing may cause a collision. By looking at the hierarchical logic of first looking at the distance and then the difficulty, the limitations of single indicator judgment are avoided, making the decision more comprehensive and safe.

[0047] In the present invention, for obstacles with sufficient spacing (such as a tracking robot traveling at a constant speed in the distance), the original path is directly used to pass, avoiding unnecessary detours or waiting, thereby improving passage efficiency. For obstacles with insufficient spacing but easy to bypass (such as slowly moving shelves with sufficient space on both sides), the shortest path is used to bypass them, which neither affects efficiency nor avoids task delays caused by waiting. For obstacles with insufficient spacing and high difficulty to bypass (such as a robotic arm moving back and forth at high speed and with a narrow space around it), the choice is to wait to pass, ensuring safety and avoiding equipment damage or collision due to forced detours.

[0048] See also Figure 4 As shown, it is a flow chart for determining the passage mode of a robot with irregular movements.

[0049] Specifically, in step S2, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the running trajectory of the irregular motion obstacle and the obstacle's running uncertainty characterization value, including: If the real-time relative distance is greater than the preset real-time relative distance, it is determined that the robot is traveling along the original driving path; If the real-time relative distance is less than the preset real-time relative distance and the obstacle's operational uncertainty representation value is less than or equal to the preset operational uncertainty representation value, then it is determined that the robot is bypassing the obstacle; If the real-time relative distance is smaller than the preset real-time relative distance and the obstacle's operational uncertainty characterization value is larger than the preset operational uncertainty characterization value, it is determined that the robot is waiting to pass.

[0050] Specifically, in this embodiment, the preset real-time relative distance is determined by the following method: Taking a warehouse logistics robot (running in the warehouse shelf channel, mainly avoiding forklifts and personnel with irregular movements) as an example: the robot's own safe braking distance is 5m (at a speed of 3m / s, the braking acceleration is 2m / s 2 , braking time 1.5s, calculated braking distance = 3×1.5-0.5×2×1.5 2 = 3.75m, with a safety margin of 5m. Estimated maximum range of motion for irregular obstacles (people): The maximum possible lateral / vertical movement distance of a person in a short period of time (2s) is 3m (walking speed 1.5m / s × 2s). Robot reaction delay: Within 0.5s, an obstacle may move an additional 0.75m (1.5m / s × 0.5s). Preset real-time relative distance = safe braking distance + maximum range of motion of the obstacle + displacement within reaction delay = 5m + 3m + 0.75m = 8.75m.

[0051] Specifically, in this embodiment, the preset operational uncertainty characterization value is determined in the following manner: taking a shopping mall service robot (mainly used to avoid pedestrians with irregular movements) as an example: the maximum direction change rate of pedestrians in the shopping mall is statistically 30° / s (such as turning suddenly), and considering safety redundancy (+20%), the preset direction change rate threshold is set to 36° / s; the maximum speed change rate of pedestrians is 0.8m / s 2 (such as sudden acceleration or deceleration), safety redundancy (+20%), the preset speed change rate threshold is set to 0.96m / s 2 ; Pedestrian direction changes have a greater impact on detours, so the direction change rate weight = 0.6, speed change rate weight = 0.4; preset operation uncertainty representation value = (36° / s×0.6) + (0.96m / s 2 ×0.4)=21.6+0.384=21.984.

[0052] Specifically, the determination of the operational uncertainty characterization value of the obstacle includes: Calculate the ratio of the obstacle's direction change rate to a preset direction change rate to obtain a first operational uncertainty sub-parameter. Calculate the ratio of the speed change rate of the obstacle to the preset speed change rate to obtain the second operation uncertainty sub-parameter, The first operational uncertainty sub-parameter and the second operational uncertainty sub-parameter are weightedly summed to obtain an operational uncertainty characterization value.

[0053] Specifically, in this embodiment, the preset direction change rate is determined in the following manner. Taking pedestrians as an example, pedestrians change direction smoothly during normal walking, but the maximum direction change when suddenly avoiding or turning is approximately 45° / s (for example, a 45° deflection from forward to side within 1 second). Considering the sensor direction detection error of ±3° and the possible additional deflection angle of the pedestrian within the robot's 0.2-second response delay (45° / s×0.2s=9°), the total redundancy is 12°. The preset direction change rate is calculated as follows: typical value + safety margin = 45° / s + 12° / s = 57° / s (rounded to 60° / s).

[0054] Specifically, the determination of the real-time relative distance includes: Collect robot position data and obstacle position data in real time, and calculate the spatial straight-line distance as the initial real-time relative distance based on the robot position data and the obstacle position data; Obtain obstacle movement direction and speed data, and calculate the relative distance change per unit time; The initial real-time relative distance is corrected based on the relative distance change to obtain the real-time relative distance.

[0055] Specifically, in step S3, determining the range of the motion trajectory includes: extracting a position coordinate sequence of the obstacle based on the image data preprocessed in step S1 to form a discrete motion trajectory point set; Obtain the maximum and minimum displacements of each coordinate axis to obtain the boundary points; Use smooth curves to connect each boundary point to obtain the range of motion trajectory; The detour trajectories with the shortest distance and the least difficulty based on the robot's position include: Determine the starting point and end point based on the robot's real-time position coordinates and the target coordinates; Generate several candidate detour trajectories that meet the constraints; The detour difficulties of the candidate trajectories are evaluated, and the candidate trajectory with the minimum detour difficulty is selected as the detour trajectory.

[0056] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent navigation method for a robot, characterized in that: include: Step S1, obtaining an obstacle image in the robot's running trajectory in real time, preprocessing the obtained obstacle image, extracting the obstacle's motion characteristics, and determining the obstacle's motion type based on the obstacle's motion characteristics; Step S2, determining the robot's obstacle avoidance method based on the determined movement type of the obstacle, includes: Determine the robot's passage method based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle; Alternatively, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the trajectory of the irregularly moving obstacle and the obstacle's operational uncertainty characterization value; Step S3, when the passage mode is determined to be bypassing, analyzing the movement trajectory of the irregularly moving obstacle to determine the range of the movement trajectory, and selecting the bypass trajectory with the shortest bypass distance and the least difficulty based on the robot position; Step S4, determining a travel trajectory according to the determined travel mode; Determining the operational uncertainty characterization value of the obstacle includes: Calculate the ratio of the obstacle's direction change rate to a preset direction change rate to obtain a first operational uncertainty sub-parameter. Calculate the ratio of the speed change rate of the obstacle to the preset speed change rate to obtain the second operation uncertainty sub-parameter, The first operational uncertainty sub-parameter and the second operational uncertainty sub-parameter are weightedly summed to obtain an operational uncertainty characterization value.

2. The intelligent navigation method for a robot according to claim 1, characterized in that: In step S1, determining the movement type of the obstacle based on the movement characteristics of the obstacle includes: Based on the pre-processed obstacle image, the motion feature data of the obstacle at several consecutive time nodes are extracted according to a predetermined interval to obtain a time series feature sequence; Analyzing the motion feature data in the time series feature sequence and calculating core feature data; Compare the core feature data with the preset thresholds respectively; Based on the comparison result, it is determined whether there is irregular core feature data, and when irregular core feature data exists, the movement type of the obstacle is determined to be irregular movement.

3. The intelligent navigation method for a robot according to claim 2, characterized in that: The core feature data includes speed standard deviation, trajectory motion deviation and direction change frequency.

4. The intelligent navigation method for a robot according to claim 2, characterized in that: In step S2, determining the robot's obstacle avoidance method based on the determined movement type of the obstacle includes: If the obstacle's motion type is regular, the robot's passage method is determined based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle. If the movement type of the obstacle is irregular movement, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the movement trajectory of the irregularly moving obstacle and the obstacle's movement uncertainty characterization value.

5. The intelligent navigation method for a robot according to claim 4, characterized in that: In step S2, the difficulty of the robot circumventing the obstacle is determined as follows: Determine the maximum change in the actual traversable space when the robot circumvents the obstacle, Determine the angle between the obstacle's running direction and the robot's current running direction, and obtain the curvature of the detour path. Calculate the ratio of the maximum change in the actual traversable space to the preset maximum change to obtain the first sub-detour difficulty parameter. Calculate the ratio of the detour path curvature to the preset detour path curvature to obtain the second sub-detour difficulty parameter, The first bypass difficulty parameter and the second sub-bypass difficulty parameter are weighted and summed to obtain the bypass difficulty.

6. The intelligent navigation method for a robot according to claim 5, characterized in that: In step S2, the robot's passage mode is determined based on the relative motion distance between the obstacle and the robot and the difficulty of the robot circumventing the obstacle, including: If the relative motion distance is greater than or equal to the preset relative motion distance, it is determined that the robot is traveling along the original driving path; If the relative motion distance is less than the preset relative motion distance and the detour difficulty is less than or equal to the preset detour difficulty, it is determined that the robot has detoured; If the relative motion distance is smaller than the preset relative motion distance and the detour difficulty is greater than the preset detour difficulty, it is determined that the robot is waiting to pass.

7. The intelligent navigation method for a robot according to claim 4, characterized in that: In step S2, the robot's passage mode is determined based on the real-time relative distance when the robot reaches the running trajectory of the irregular motion obstacle and the obstacle's running uncertainty characterization value, including: If the real-time relative distance is greater than the preset real-time relative distance, it is determined that the robot is traveling along the original driving path; If the real-time relative distance is less than the preset real-time relative distance and the obstacle's operational uncertainty representation value is less than or equal to the preset operational uncertainty representation value, then it is determined that the robot is bypassing the obstacle; If the real-time relative distance is smaller than the preset real-time relative distance and the obstacle's operational uncertainty characterization value is larger than the preset operational uncertainty characterization value, it is determined that the robot is waiting to pass.

8. The intelligent navigation method for a robot according to claim 7, characterized in that: The determination of the real-time relative spacing includes: Collect robot position data and obstacle position data in real time, and calculate the spatial straight-line distance as the initial real-time relative distance based on the robot position data and the obstacle position data; Obtain obstacle movement direction and speed data, and calculate the relative distance change per unit time; The initial real-time relative distance is corrected based on the relative distance change to obtain the real-time relative distance.

9. The intelligent navigation method for a robot according to claim 1, characterized in that: In step S3, determining the range of the motion trajectory includes: Based on the image data preprocessed in step S1, the position coordinate sequence of the obstacle is extracted to form a discrete motion trajectory point set; Obtain the maximum and minimum displacements of each coordinate axis to obtain the boundary points; Use smooth curves to connect each boundary point to obtain the range of motion trajectory; The detour trajectories with the shortest distance and the least difficulty based on the robot's position include: Determine the starting point and end point based on the robot's real-time position coordinates and the target coordinates; Generate several candidate detour trajectories that meet the constraints; The detour difficulties of the candidate trajectories are evaluated, and the candidate trajectory with the minimum detour difficulty is selected as the detour trajectory.

Images