Inspection path planning methods and inspection robots

Abstract

This application discloses an inspection path planning method and an inspection robot, which are applied to the inspection robot. The method includes: during the process of performing an inspection task based on the original path, collecting motion trajectory data of obstacles at multiple moments within a preset sliding time window, the motion trajectory data including the direction angle and velocity at each moment; determining the motion type of the obstacle based on the collected motion trajectory data; predicting the motion path of the obstacle within a preset time period based on the motion type of the obstacle and the collected motion trajectory data; and adjusting the original path based on the predicted motion path.

Description

Technical Field

[0001] This application relates to the field of automatic control technology, and in particular to an inspection path planning method and an inspection robot. Background Technology

[0002] With the modernization and automation of power systems, the daily inspection of substations, as crucial power facilities, is of paramount importance to ensuring stable equipment operation. Traditional inspection methods rely on manual operation, which, while allowing for relatively direct checks, involves high workload and safety risks, is inefficient, and is easily affected by environmental factors. With technological advancements, inspection robots have gradually become a key technology for improving inspection efficiency and reducing labor costs and risks.

[0003] However, the substation environment is full of obstacles, which may come from various factors such as power equipment, pipelines, pedestrians, and vehicles. This requires inspection robots to have a high degree of flexibility and intelligent obstacle avoidance capabilities when performing tasks, in order to cope with complex and ever-changing environments. Although existing inspection robot path planning technologies are usually based on environmental perception information for path design, these methods are still insufficient in dealing with dynamic environments.

[0004] In particular, while existing technologies are relatively mature in handling static obstacles, they lag behind in handling moving obstacles. Traditional inspection path planning fails to fully consider the impact of dynamic factors on obstacles, which may lead to path planning errors by the inspection robot during obstacle avoidance, resulting in the risk of collisions and affecting the smooth progress of the inspection task.

[0005] Improving the ability of inspection robots to recognize obstacles of various motion types, achieving precise planning of inspection paths, and reducing the risk of obstacle collisions are key issues that need to be addressed. Summary of the Invention

[0006] The purpose of this application is to provide an inspection path planning method and robot to solve the problem of low path planning accuracy in complex and varied environmental obstacles caused by tire blowouts.

[0007] To solve the above-mentioned technical problems, this specification is implemented as follows: Firstly, an inspection path planning method is provided for use with an inspection robot, the method comprising: During the inspection task movement based on the original path, the motion trajectory data of obstacles at multiple moments within a preset sliding time window are collected. The motion trajectory data includes the direction angle and velocity at each moment. Based on the collected motion trajectory data, the motion type of the obstacle is determined; Based on the movement type of the obstacle and the collected movement trajectory data, predict the movement path of the obstacle within a preset time period; The original path is adjusted based on the predicted motion path.

[0008] Optionally, based on the collected motion trajectory data, the motion type of the obstacle is determined, including: Based on the direction angles at multiple moments in the collected motion trajectory data, the standard deviation of the direction angles is calculated. Based on the velocity at multiple moments in the collected motion trajectory data, the mean velocity, standard deviation of velocity, and mean acceleration are calculated respectively. The motion type of the obstacle is determined based on the standard deviation of the orientation angle, the mean velocity, the standard deviation of the velocity, and the mean acceleration.

[0009] Optionally, the motion type of the obstacle is determined based on the standard deviation of the orientation angle, the mean velocity, the standard deviation of the velocity, and the mean acceleration, including: If the standard deviation of the orientation angle is less than a first threshold, the standard deviation of the velocity is less than a second threshold, and the mean acceleration is less than a third threshold, then the motion type of the obstacle is determined to be uniform motion; or If the standard deviation of the orientation angle is less than the first threshold, the standard deviation of the velocity is less than the second threshold, and the mean acceleration is not less than the third threshold, then the motion type of the obstacle is determined to be a regular motion of accelerated motion; or If the standard deviation of the orientation angle is not less than the first threshold or the standard deviation of the velocity is not less than the second threshold, then the motion type of the obstacle is determined to be irregular motion; or If the direction angle and velocity at the multiple moments are both 0, then the motion type of the obstacle is determined to be stationary.

[0010] Optionally, predicting the movement path of the obstacle within a preset time period based on the obstacle's movement type and the collected movement trajectory data includes: If the obstacle's motion type is uniform motion, then the mean value of the direction angle is calculated based on the direction angles at multiple moments in the collected motion trajectory data; Calculate the average speed based on the speed at multiple moments in the collected motion trajectory data; Calculate the first moving distance corresponding to the preset duration based on the average speed; The movement path of the obstacle within a preset time period is predicted based on the mean azimuth angle and the first moving distance.

[0011] Optionally, predicting the movement path of the obstacle within a preset time period based on the obstacle's movement type and the collected movement trajectory data includes: If the obstacle's motion type is accelerated motion, then the mean value of the direction angle is calculated based on the direction angles at multiple moments in the collected motion trajectory data; Based on the velocity at multiple moments in the collected motion trajectory data, the mean velocity and mean acceleration are calculated respectively; The second moving distance corresponding to the preset duration is calculated based on the average velocity and the average acceleration. Based on the mean azimuth angle and the second moving distance, the movement path of the obstacle within a preset time period is predicted.

[0012] Optionally, predicting the movement path of the obstacle within a preset time period based on the obstacle's movement type and the collected movement trajectory data includes: If the obstacle's motion type is irregular, then the median velocity is calculated based on the velocity at multiple moments in the collected motion trajectory data; Calculate the third travel distance corresponding to the preset duration based on the median speed; The second obstacle avoidance length is obtained by multiplying the third moving distance and the preset safety factor. A circle is drawn with the center of the plane where the obstacle is currently located as the center and the second obstacle avoidance length as the radius to obtain the obstacle avoidance area; The obstacle avoidance area is predicted as the movement path of the obstacle within a preset time period.

[0013] Optionally, predicting the movement path of the obstacle within a preset time period based on the obstacle's movement type and the collected movement trajectory data includes: If the obstacle's motion type is stationary, then obtain the first intersection point where the obstacle's current position plane intersects with the original path; The first obstacle avoidance length is determined based on the length of the obstacle and the length of the inspection robot; With the first intersection point as the center and the third obstacle avoidance length as the radius, a circle is drawn to obtain the obstacle avoidance area; The obstacle avoidance area is predicted as the movement path of the obstacle within a preset time period.

[0014] Optionally, adjusting the original path based on the predicted motion path includes: Determine the second intersection point between the predicted motion path and the original path; The first obstacle avoidance length is determined based on the length of the obstacle and the length of the inspection robot; With the second intersection point as the center and the first obstacle avoidance length as the radius, draw a circle to obtain the obstacle avoidance area; Obtain the third and fourth intersection points where the obstacle avoidance area intersects with the original path; The original path between the third and fourth intersection points is replaced with the outer edge of the first obstacle avoidance area between the third and fourth intersection points to obtain the adjusted path.

[0015] Optionally, adjusting the original path based on the predicted motion path includes: Obtain the fifth and sixth intersection points where the obstacle avoidance area intersects with the original path; The original path between the fifth and sixth intersection points is replaced with the outer edge of the obstacle avoidance area between the fifth and sixth intersection points to obtain the adjusted path.

[0016] In a second aspect, an inspection robot is provided, including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method described in the first aspect.

[0017] In this embodiment, the inspection robot collects motion trajectory data of obstacles at multiple moments within a preset sliding time window during the process of moving based on the original path to perform inspection tasks. The motion trajectory data includes the direction angle and velocity at each moment. Based on the collected motion trajectory data, the robot determines the motion type of the obstacle. Based on the motion type of the obstacle and the collected motion trajectory data, the robot predicts the motion path of the obstacle within a preset time period. Based on the predicted motion path, the robot adjusts the original path. Therefore, compared with traditional path planning methods, this embodiment can dynamically calculate the motion type of the obstacle and the possible future motion path of the obstacle based on the collected motion trajectory data, and make adjustments in advance according to the predicted motion type and motion path. This significantly improves the inspection robot's ability to identify obstacles of various motion types, achieves accuracy and real-time performance in inspection path planning and obstacle avoidance, reduces the risk of obstacle collisions, ensures the safety and efficiency of path planning, and improves the intelligence level and adaptability of the inspection robot in dynamic environments. Especially in complex environments, path planning needs to be dynamically adjusted in real time. By fully considering the impact of dynamic factors of obstacles, timely adjustment of the original path can effectively avoid unnecessary detours and obstacle contact, making path planning more efficient and reducing path adjustment time and energy consumption. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart illustrating the inspection path planning method according to an embodiment of this application.

[0019] Figure 2 This is a schematic diagram of the original path adjustment scenario in an embodiment of this application.

[0020] Figure 3 This is a structural block diagram of the inspection robot according to an embodiment of this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. The drawing numbers in this application are only used to distinguish the various steps in the solution and are not used to limit the execution order of the various steps. The specific execution order is subject to the description in the specification.

[0022] To address the problems existing in the prior art, this application provides an inspection path planning method applied to an inspection robot, such as... Figure 1 As shown, it includes the following steps 102 to 108.

[0023] Step 102: During the inspection task movement based on the original path, collect the motion trajectory data of the obstacles at multiple moments within a preset sliding time window. The motion trajectory data includes the direction angle and speed at each moment.

[0024] During the inspection robot's movement while performing its inspection tasks, it collects real-time motion trajectory data of obstacles in the surrounding environment. The motion trajectory data includes the direction angle and velocity at various times. Obstacles can include moving objects such as pedestrians and vehicles, as well as stationary objects such as electrical equipment and pipelines. The corresponding motion trajectory data will show different values.

[0025] Here, obstacles refer to objects that do not appear in the preset map corresponding to the original path. The inspection robot collects the positions of identifiable objects in the surrounding area and compares them with the positions of objects in the preset map. If the object does not exist in the preset map, the corresponding object is identified as an obstacle.

[0026] When collecting motion trajectory data, the inspection robot maintains a data buffer of fixed length (or fixed duration) (i.e., a preset sliding time window). It continuously stores the motion trajectory data of each identified obstacle at multiple moments within a recent period. The motion trajectory data collected at each moment can be represented as: time t - direction angle α - velocity v.

[0027] Each time motion trajectory data point is collected, it is added to the end of the sliding time window. If the sliding time window is full (reaching the preset maximum capacity), the earliest data point collected at the beginning of the window is discarded. All data points within the entire sliding time window are used in subsequent steps to determine the motion type of the corresponding obstacle and predict its future path. The size of the sliding time window can range from, for example, 0.5 to 3 seconds.

[0028] Step 104: Determine the motion type of the obstacle based on the collected motion trajectory data.

[0029] As mentioned above, obstacles can include moving obstacles and / or stationary obstacles, and the corresponding motion types include regular uniform motion, regular accelerated motion, irregular motion, and stationary motion. Different motion types can be determined based on motion trajectory data collected at multiple moments within a preset sliding time window, including the direction angle and velocity at each moment.

[0030] Based on the solution provided in the above embodiments, optionally, in step 104 above, determining the motion type of the obstacle based on the collected motion trajectory data includes: calculating the standard deviation of the direction angle based on the direction angle at multiple moments in the collected motion trajectory data; calculating the mean velocity, the standard deviation of velocity, and the mean acceleration based on the velocity at multiple moments in the collected motion trajectory data; and determining the motion type of the obstacle based on the standard deviation of the direction angle, the mean velocity, the standard deviation of velocity, and the mean acceleration.

[0031] Based on the obstacle's motion trajectory data, its motion pattern needs to be classified to predict its future path. Uniform motion refers to the obstacle moving at a constant speed along a fixed direction (e.g., automated transport vehicles, patrol robots). Accelerated motion refers to the obstacle moving with a constant acceleration in a certain direction (e.g., a vehicle accelerating). Irregular motion refers to the obstacle's irregular motion, making its specific path difficult to predict (e.g., the random movement of a person or animal). Stationary motion refers to the obstacle remaining in a fixed position (e.g., a parked vehicle or other stationary object).

[0032] To determine the direction of an obstacle's movement, it is necessary to calculate the direction angle between corresponding points on the obstacle at two adjacent moments. The direction angle is the angle between the line connecting the two points and a reference direction (such as a horizontal line). Calculating the direction angle helps to understand whether the obstacle's direction of movement is stable or has changed drastically.

[0033] By statistically analyzing the direction angles at multiple moments within the motion trajectory data collected during the current preset sliding time window, the standard deviation of the direction angles at each adjacent moment can be calculated. Standard deviation is an indicator of data fluctuation; here, the standard deviation of the direction angle represents the fluctuation in the movement direction of the same obstacle. If the fluctuation in the direction angle is small (small standard deviation), it indicates that the obstacle's movement direction is relatively stable. If the standard deviation is large, it indicates that the obstacle's movement direction changes significantly, possibly indicating a sharp turn.

[0034] By statistically analyzing the velocity at multiple moments in the motion trajectory data collected within the current preset sliding time window, the mean velocity, standard deviation of velocity, and standard deviation of acceleration can be calculated.

[0035] Velocity represents the magnitude of an obstacle's speed at each instant, while acceleration is the rate of change of velocity at a given instant compared to the velocity at the previous instant, indicating the trend of the obstacle's motion. By calculating the changes in velocity over multiple adjacent instants, the mean of acceleration can be obtained. The mean of acceleration can help determine whether the obstacle is moving at a constant speed or accelerating. If the acceleration is small or close to zero, the obstacle is likely moving at a constant speed; if the acceleration is large, it is likely accelerating.

[0036] Therefore, based on the calculated standard deviation of the direction angle, the mean velocity, the standard deviation of the velocity, and the mean acceleration, the motion type of the obstacle can be determined.

[0037] In one specific embodiment, the motion type of the obstacle is determined based on the standard deviation of the direction angle, the mean velocity, the standard deviation of the velocity, and the mean acceleration, including: if the standard deviation of the direction angle is less than a first threshold, the standard deviation of the velocity is less than a second threshold, and the mean acceleration is less than a third threshold, then the motion type of the obstacle is determined to be regular uniform motion; or if the standard deviation of the direction angle is less than the first threshold, the standard deviation of the velocity is less than the second threshold, and the mean acceleration is not less than the third threshold, then the motion type of the obstacle is determined to be regular accelerated motion; or if the standard deviation of the direction angle is not less than the first threshold or the standard deviation of the velocity is not less than the second threshold, then the motion type of the obstacle is determined to be irregular motion; or if the direction angle and velocity at multiple moments are all 0, then the motion type of the obstacle is determined to be stationary.

[0038] It is worth noting that the first threshold (corresponding to the standard deviation of the direction angle) is used to determine whether the direction of obstacle movement is regular. The smaller the standard deviation of the direction angle, the more stable the direction. Regular movements (such as vehicles traveling along a road) usually have very small fluctuations in the direction angle, with a standard deviation of less than 10 degrees. Irregular movements (such as people walking or animals running) may change direction frequently, and the standard deviation can easily exceed 20-30 degrees.

[0039] Setting the threshold too low (e.g., 5 degrees) may misinterpret some slightly swaying, regular movements as irregular movements, leading to overly conservative obstacle avoidance. Setting it too high (e.g., 45 degrees) may misinterpret the movements of people with large directional changes as regular movements, resulting in risks associated with precise prediction. In this embodiment, the first threshold value ranges from, for example, between 10 and 30 degrees. Optionally, the first threshold can be set to 15 degrees, which can better distinguish between driving along a road and wandering around randomly.

[0040] The second threshold (corresponding to the speed standard deviation threshold) is used to determine whether the obstacle's movement speed is stable. The speed standard deviation reflects whether the obstacle is moving at a constant speed or fluctuating. In regular motion, even accelerated motion, the speed change is orderly, and the speed standard deviation is low (for example, the speed standard deviation of uniformly accelerated motion is theoretically 0, but actual sensor data will have slight fluctuations). In irregular motion, the speed changes erratically, and the standard deviation is large. In this embodiment, the second threshold ranges from 0.2 m / s to 0.5 m / s. Optionally, the second threshold can be set to 0.3 m / s, which can effectively capture the transition of a pedestrian from walking to running, while tolerating slight fluctuations during smooth vehicle acceleration.

[0041] The third threshold (corresponding to the threshold of the average acceleration) is used to distinguish between uniform motion and accelerated motion. The absolute value of the average acceleration reflects the degree of speed change; close to 0 indicates uniform motion, while significantly greater than 0 indicates accelerated or decelerated motion. The third threshold is set to filter out spurious accelerations caused by sensor noise or small speed fluctuations. In a substation environment, the normal start-stop acceleration of inspection vehicles is typically greater than 0.5 m / s², while the acceleration variation of personnel walking is smaller. In this embodiment, the value range includes 0.1 m / s² to 0.3 m / s². Optionally, the third threshold can be set to 0.15 m / s², which can effectively distinguish between basically uniform movement and obvious acceleration / deceleration states.

[0042] Therefore, based on the motion trajectory data collected within the aforementioned preset sliding time window, the standard deviation of the direction angle, the mean velocity, the standard deviation of the velocity, and the mean acceleration are calculated to determine the current motion type of the obstacle.

[0043] Specifically, if the standard deviation of the direction angle is less than the first threshold (i.e., the direction angle fluctuates little and the direction of movement is stable), the standard deviation of the velocity is less than the second threshold (i.e., the velocity changes little and the velocity of movement is stable), and the mean value of the acceleration is less than the third threshold (i.e., the acceleration is small and the velocity of movement is uniform), then the motion type corresponding to the obstacle is determined to be a regular motion of uniform velocity.

[0044] If the standard deviation of the orientation angle is less than the first threshold (i.e., the orientation angle fluctuates little and the direction of movement is stable), the standard deviation of the velocity is less than the second threshold (i.e., the velocity changes little and the velocity is stable), and the mean acceleration is not less than the third threshold (i.e., the acceleration is large, indicating that the obstacle is accelerating), then the motion type corresponding to the obstacle is determined to be a regular motion of accelerated movement.

[0045] If the standard deviation of the orientation angle is not less than the first threshold (i.e., the orientation angle fluctuates greatly) or the standard deviation of the velocity is not less than the second threshold (i.e., the velocity fluctuates greatly), then the motion type corresponding to the obstacle is determined to be irregular motion.

[0046] If the direction angle and velocity collected within the preset sliding time window are both 0, it indicates that the obstacle is not moving and is stationary. In this case, the motion type of the obstacle is directly determined to be stationary.

[0047] By introducing the mean and standard deviation analysis of orientation angle, velocity, and acceleration, the classification of obstacle motion types achieves high accuracy and stability. The calculation of standard deviation effectively suppresses accidental abnormal fluctuations, ensuring accurate identification of obstacle motion patterns under different environments. For obstacles with relatively regular motion patterns, it can accurately capture their motion trends, providing reliable data support for path adjustment.

[0048] By meticulously analyzing real-time obstacle trajectory data, complex obstacle motion characteristics are abstracted into quantifiable parameters, which are then automatically categorized based on preset thresholds. This data-driven motion type discrimination method not only avoids manual intervention and improves automation, but also enables real-time updates and adjustments to obstacle motion category recognition results in dynamic environments, enhancing the system's intelligence.

[0049] Step 106: Based on the movement type of the obstacle and the collected movement trajectory data, predict the movement path of the obstacle within a preset time period.

[0050] For different motion types, the predicted movement path of the obstacle within a preset time period will differ. The preset time period is the length of time it takes to predict the future location of the obstacle, i.e., how far it can be seen. The movement path within the preset time period refers to the possible movement path of the obstacle within the preset time period from the moment the obstacle's motion type is determined, based on the motion trajectory data collected in the current preset sliding time window.

[0051] The lower limit of the preset duration needs to allow the inspection robot a minimum reaction time to perceive, calculate, and initiate obstacle avoidance actions. Less than this minimum time may prevent the inspection robot from responding to unexpected situations. The upper limit of the preset duration is based on the reliability of the prediction; for dynamically moving obstacles, the longer the prediction time, the greater the uncertainty. In this embodiment, the preset duration ranges from 1 second to 3 seconds.

[0052] In open areas or when moving objects are moving slowly, a longer preset time (e.g., 3 seconds) can be set to allow for smoother and more efficient path adjustments to begin earlier. In complex areas such as narrow corridors or intersections, or when moving objects are moving quickly, a shorter preset time (e.g., 1 second) should be set to ensure safety is the top priority and allow for more timely avoidance maneuvers.

[0053] In one specific embodiment, predicting the movement path of the obstacle within a preset time period based on the obstacle's movement type and the collected movement trajectory data includes: if the obstacle's movement type is uniform motion, calculating the average direction angle based on the direction angles at multiple moments in the collected movement trajectory data; calculating the average speed based on the speed at multiple moments in the collected movement trajectory data; calculating a first moving distance corresponding to the preset time period based on the average speed; and predicting the obstacle's movement path within the preset time period based on the average direction angle and the first moving distance.

[0054] Regular, uniform motion corresponds to obstacles with small changes in direction angle, small velocity fluctuations, and low average acceleration. This means that the obstacle's trajectory is relatively stable and regular, making it easier to predict and plan the path.

[0055] For obstacles moving at a constant speed, the path of the obstacle within a preset time period can be predicted by combining the mean direction angle and the mean velocity. Assuming the mean direction angle is... The product of the average speed and the preset time duration gives the distance traveled. The distance traveled, combined with the mean azimuth angle, yields the future position of the obstacle after a preset time. , Specifically, it can be calculated using the following formula:

[0056] here,( , () is the current position of the obstacle at the moment of prediction of the motion type. It is the mean of the direction angle. This refers to the distance traveled. Using this formula, the future position of the obstacle at a predetermined time can be calculated, and then, based on the current position (…),... , ) and future location ( , ), and the mean of the direction angle is This method can predict the motion path of obstacles moving at a constant speed. This prediction can then be used to adjust the original path, effectively avoiding collisions between the inspection robot and obstacles. This prediction method is simple and efficient, particularly suitable for obstacles moving at a stable, constant speed, making path planning adjustments smoother.

[0057] In one specific embodiment, predicting the movement path of the obstacle within a preset time period based on the obstacle's movement type and the collected movement trajectory data includes: if the obstacle's movement type is accelerated movement, calculating the average direction angle based on the direction angles at multiple moments in the collected movement trajectory data; calculating the average velocity and average acceleration based on the velocity at multiple moments in the collected movement trajectory data; calculating the second movement distance corresponding to the preset time period based on the average velocity and average acceleration; and predicting the obstacle's movement path within the preset time period based on the average direction angle and the second movement distance.

[0058] Regularly accelerating obstacles have small changes in direction angle and stable speed fluctuations, but a high average acceleration. These obstacles usually exhibit regular acceleration or deceleration characteristics, and can be effectively avoided based on the predicted acceleration trend.

[0059] The only difference between this and the predicted path of an obstacle moving at a constant speed within a preset time period is the distance the obstacle travels within that preset time period. It is calculated based on the average velocity and average acceleration, and the calculation logic for the future positions of other obstacles at corresponding moments after a preset time is the same. Therefore, the acceleration path of the obstacles can be predicted.

[0060] For obstacles with regularly accelerating motion, not only the average velocity but also the average acceleration is considered to predict the obstacle's future path. By calculating the distance traveled based on the average acceleration and velocity, combined with the average direction angle, the accelerated path of the obstacle within a preset time period can be accurately predicted. This prediction method can then be used to adjust the original path, handling changes such as obstacle acceleration or deceleration, ensuring that the inspection robot is prepared for obstacle avoidance in complex dynamic environments.

[0061] By comprehensively analyzing multiple factors such as direction angle, velocity, and acceleration, it is possible to predict the future trajectory of obstacles with regular motion with high accuracy.

[0062] In one specific embodiment, predicting the movement path of the obstacle within a preset time period based on the obstacle's movement type and the collected movement trajectory data includes: if the obstacle's movement type is irregular, calculating the median speed based on the speed at multiple moments in the collected movement trajectory data; calculating a third movement distance corresponding to the preset time period based on the median speed; obtaining a second obstacle avoidance length based on the product of the third movement distance and a preset safety factor; drawing a circle with the center of the obstacle's current position plane as the center and the second obstacle avoidance length as the radius to obtain an obstacle avoidance area; and predicting the obstacle avoidance area as the obstacle's movement path within the preset time period.

[0063] Irregular motion corresponds to a large standard deviation in direction angle and velocity. The motion of obstacles with irregular movements (such as pedestrians and animals) has no obvious pattern and may manifest as sudden acceleration, deceleration, sharp turns, etc. Such obstacles have strong uncertainty, so it is not easy to predict their future movement path using linear models corresponding to regular motion. Inspection robots need to adopt more flexible path planning strategies to cope with their unpredictable motion.

[0064] Since its movement path cannot be accurately predicted, the movement trajectory area of ​​the obstacle can be inferred from the distribution of the collected obstacle movement trajectory data in this embodiment of the application. That is, the range in which the obstacle may appear. This area needs to be avoided by the inspection robot and is referred to as the obstacle avoidance area in this article.

[0065] The following describes the specific logic for predicting the movement path of obstacles within a preset time period.

[0066] For obstacles with irregular movement, the median velocity is first extracted from the motion trajectory data collected within the current preset sliding time window across multiple moments. Velocity may fluctuate significantly across these moments, especially when the obstacle moves irregularly. Taking the median velocity effectively suppresses the influence of extreme velocity values ​​(such as abnormal velocities caused by sudden acceleration or deceleration), resulting in a smoother and more reliable velocity estimation. The median is the value in the middle when the data is arranged in ascending order.

[0067] By multiplying the extracted median velocity by a preset time duration, the distance the obstacle travels within that time duration can be obtained. This distance is then multiplied by a preset safety factor, also known as a redundancy factor, to obtain the obstacle avoidance length. The redundancy factor is a safety factor used to ensure that the obstacle avoidance area corresponding to the subsequently predicted obstacle avoidance length is large enough to prevent sudden changes in the obstacle during its movement. The introduction of the redundancy factor increases the safety of path planning, enhances the robustness and fault tolerance of the prediction results, and avoids collisions caused by calculation errors or sudden behaviors of dynamic obstacles. A reasonable setting of the redundancy factor ensures that the inspection robot can reserve sufficient obstacle avoidance space in dynamic environments, thereby guaranteeing more reliable and safer path planning.

[0068] A circular region is formed with the center of the plane representing the obstacle's current position as the center and the obstacle avoidance length as the radius. The plane representing the obstacle's current position refers to the plane occupied or projected by the obstacle's current location. For example, if the obstacle is a vehicle, the plane representing the obstacle's current position is the area projected onto the ground from the vehicle's corresponding location. The center of the circular region is the obstacle's current position, and the radius is calculated based on the distance traveled. This means that the obstacle's future path will be represented as a circular region with the current obstacle position as the center and the obstacle avoidance length as the radius. This circular region represents the obstacle's possible range of movement over a predetermined time period. The inspection robot needs to avoid this region to prevent collisions.

[0069] Understandably, since the direction of obstacle movement is unpredictable during irregular motion, it is necessary to define an obstacle avoidance zone to accommodate the possible directions and distances of obstacle movement. Subsequently, during path planning, the inspection robot will adjust its route based on this obstacle avoidance zone to ensure it bypasses the area and avoids collisions with obstacles.

[0070] In this embodiment, by extracting the median velocity from the currently collected motion trajectory data, interference from abnormal data can be effectively eliminated, ensuring the stability and representativeness of the velocity calculation. Determining the median velocity allows for a more accurate prediction of the obstacle's movement trend in the future, especially under irregular movement conditions, identifying the possible movement area covered by the obstacle and providing more reliable trajectory prediction for path planning. This prediction method can flexibly adapt to trajectory changes caused by irregular movement, accurately predict the possible movement range of the obstacle, and prevent path planning deviations caused by movement uncertainties.

[0071] By using a circular area as the motion path, an intuitive, concise, and highly reliable reference can be provided for the robot to plan obstacle avoidance areas. The circular area can cover the possible future positions of obstacles, ensuring that the robot can fully consider the range of movement of obstacles and make obstacle avoidance decisions in advance. This effectively avoids inaccurate path planning or collision risks caused by the irregular movement of obstacles, improving the safety and obstacle avoidance capabilities of inspection robots in complex dynamic environments.

[0072] In one specific embodiment, predicting the movement path of the obstacle within a preset time period based on the movement type of the obstacle and the collected movement trajectory data includes: if the movement type of the obstacle is stationary, obtaining the first intersection point where the current position plane of the obstacle intersects with the original path; determining a first obstacle avoidance length based on the length of the obstacle and the length of the inspection robot; drawing a circle with the first intersection point as the center and the third obstacle avoidance length as the radius to obtain an obstacle avoidance area; and predicting the obstacle avoidance area as the movement path of the obstacle within the preset time period.

[0073] A stationary obstacle has both its direction angle and velocity at zero, indicating that it has not moved at all within the current preset sliding time window. The obstacle's path prediction within the preset time period can be achieved by drawing a circular area with the intersection of the obstacle's stationary position plane and the original path as the center, and the obstacle avoidance distance determined by the obstacle's length and the inspection robot's length as the radius. This circular area represents the fixed range of the obstacle's movement within a certain time period corresponding to the preset time in the future. The inspection robot must avoid this area to prevent collisions.

[0074] Obstacle length refers to the length in the obstacle's dimensional data, which is usually considered in path planning. Inspection robot length refers to the length in the inspection robot's own dimensional data.

[0075] To ensure that the inspection robot has enough space to bypass obstacles, an additional avoidance redundancy space is usually preset in the path planning.

[0076] The formula for calculating obstacle avoidance length is as follows:

[0077] in, It is the obstacle avoidance length. It is the length of the obstacle. It is the length of the inspection robot. It avoids redundancy.

[0078] Understandably, a stationary obstacle has a certain length and occupies a certain projected area on a plane. An obstacle avoidance zone needs to be defined to accommodate the obstacle. The purpose of this zone is to ensure that the obstacle will not cause a direct collision with the inspection robot; within this zone, the robot must avoid the obstacle. Subsequently, in path planning, the inspection robot will adjust its route based on this avoidance zone to ensure it bypasses the area and avoids collisions with obstacles.

[0079] It's important to note that the "stationary" motion type of the obstacle here only indicates the current determination based on the collected motion trajectory data. If the obstacle moves again later, the motion type will be reassessed based on step 104 above, determining whether it's regular or irregular motion. For example, if a vehicle that was initially stationary starts moving, its motion type will change.

[0080] In this embodiment, by obtaining the intersection of the plane where the obstacle is located and the original path of the inspection robot, and combining it with the obstacle avoidance length, the area covered or occupied by the obstacle in the future period can be predicted more accurately, providing a more reliable trajectory prediction for path planning.

[0081] Step 108: Adjust the original path based on the predicted motion path.

[0082] As mentioned above, motion types include uniform motion, accelerated motion, irregular motion, and stillness.

[0083] For regular motion including uniform motion and accelerated motion, in one specific embodiment, adjusting the original path based on the predicted motion path includes: determining a second intersection point where the predicted motion path intersects with the original path; determining a first obstacle avoidance length based on the length of the obstacle and the length of the inspection robot; drawing a circle with the second intersection point as the center and the first obstacle avoidance length as the radius to obtain an obstacle avoidance area; obtaining a third and fourth intersection point where the obstacle avoidance area intersects with the original path; replacing the original path between the third and fourth intersection points with the outer edge of the first obstacle avoidance area between the third and fourth intersection points to obtain the adjusted path.

[0084] Specifically, for obstacles with regular movement, the original path of the inspection robot is adjusted based on the predicted movement path of the obstacle to avoid conflict between the inspection robot and the obstacle. The logic is as follows.

[0085] The predicted path of an obstacle with regular movement is usually a straight line or a curve. If it intersects with the original path of the inspection robot at a second intersection point, a circle can be drawn with the second intersection point as the center and the obstacle avoidance length determined by the length of the obstacle and the length of the inspection robot as the radius. This can form a circular obstacle avoidance area on the original path.

[0086] The radius of the obstacle avoidance zone, i.e. the obstacle avoidance length, is calculated by taking into account both the obstacle and the robot's dimensions. This avoids redundancy and excessive avoidance in path planning, ensuring both safety and improving the effectiveness of the path.

[0087] Combination Figure 2 The original path 10 and the obstacle avoidance area 20 have two intersection points: the third intersection point and the fourth intersection point, namely the intersection point a which is closer to the inspection robot and the intersection point b which is farther away from the inspection robot on the circumference of the obstacle avoidance area. Intersection points a and b divide the outer edge of the obstacle avoidance area 20, i.e. the circumference edge, into arc 22 and arc 24.

[0088] The adjusted path is obtained by replacing the local original path 12 between intersection point a and intersection point b within the obstacle avoidance area 20 in the original path 10 with the circumferential edge of the obstacle avoidance area 20 between intersection point a and intersection point b.

[0089] Local original path 12 refers to the portion of the original path 10 between intersection point a and intersection point b. This portion of the path needs to be replaced to avoid interference from obstacles. If the circumferential edge of the circular obstacle avoidance area between intersection point a and intersection point b has two arcs, such as arc 22 and arc 24, and the corresponding arc lengths of the two arcs are different, such as... Figure 2 As shown, the arc length of arc 22 is shorter than that of arc 24. Therefore, the part of the circumference edge of arc 22 with the shorter arc length can be selected as the replacement path of the local original path 12.

[0090] In this embodiment, by determining the intersection point between the predicted movement path of the obstacle and the original path, the relative position of the obstacle and the original path can be accurately identified. Based on this, combined with the obstacle length, the inspection robot length, and the avoidance redundancy, a suitable obstacle avoidance length is calculated and the obstacle avoidance area is determined, providing a scientific basis for subsequent adjustments to the original path. Thus, in a dynamic environment where obstacles move in a predictable manner, collisions between obstacles and the inspection robot can be effectively avoided, improving the intelligence and safety of path planning.

[0091] Furthermore, by intelligently adjusting the local path of the original path, it is possible not only to ensure that the inspection robot avoids obstacles, but also to guarantee the continuity and smoothness of path planning, enabling the inspection robot to complete the inspection task smoothly and reducing the waste of time and energy.

[0092] For irregular motion and stillness, in one specific embodiment, the original path is adjusted based on the predicted motion path, including: obtaining the fifth and sixth intersection points of the obstacle avoidance area and the original path; replacing the original path between the fifth and sixth intersection points with the outer edge of the obstacle avoidance area between the fifth and sixth intersection points to obtain the adjusted path.

[0093] As mentioned above, for obstacles with irregular or stationary motion, the obstacle avoidance area corresponding to the obstacle can be predicted as the obstacle's motion path within a preset time period.

[0094] Therefore, given that the obstacle avoidance area is known, we directly obtain the two intersection points between the obstacle avoidance area and the original path: the fifth intersection point and the sixth intersection point, which are the intersection points on the circumference of the obstacle avoidance area that are closer to the inspection robot and the intersection points that are farther away from the inspection robot. Then, we replace the local original path corresponding to these two intersection points with the circumferential edge of the obstacle avoidance area between these two intersection points to obtain the adjusted path.

[0095] The specific path adjustment method in this embodiment is the same as the method of locally adjusting the original path by combining the two intersection points corresponding to the third and fourth intersection points in the obstacle avoidance area of ​​the above-mentioned regular movement type. For details, please refer to the above text, which will not be repeated here.

[0096] In this embodiment, the inspection robot collects motion trajectory data of obstacles at multiple moments within a preset sliding time window during the process of moving based on the original path to perform inspection tasks. The motion trajectory data includes the direction angle and velocity at each moment. Based on the collected motion trajectory data, the robot determines the motion type of the obstacle. Based on the motion type of the obstacle and the collected motion trajectory data, the robot predicts the motion path of the obstacle within a preset time period. Based on the predicted motion path, the robot adjusts the original path. Therefore, compared with traditional path planning methods, this embodiment can dynamically calculate the motion type of the obstacle and the possible future motion path of the obstacle based on the collected motion trajectory data, and make adjustments in advance according to the predicted motion type and motion path. This significantly improves the inspection robot's ability to identify obstacles of various motion types, achieves accuracy and real-time performance in inspection path planning and obstacle avoidance, reduces the risk of obstacle collisions, ensures the safety and efficiency of path planning, and improves the intelligence level and adaptability of the inspection robot in dynamic environments. Especially in complex environments, path planning needs to be dynamically adjusted in real time. By fully considering the impact of dynamic factors of obstacles, timely adjustment of the original path can effectively avoid unnecessary detours and obstacle contact, making path planning more efficient and reducing path adjustment time and energy consumption.

[0097] Optionally, such as Figure 3 As shown, this application embodiment also provides an inspection robot 2000, including a processor 2400 and a memory 2200. The memory 2200 stores a program or instructions that can run on the processor 2400. When the program or instructions are executed by the processor 2400, they implement the various steps of the above-described inspection path planning method embodiment and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0098] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of any of the above-described inspection path planning method embodiments and achieve the same technical effect. To avoid repetition, further details are omitted here. The readable storage medium includes computer-readable storage media, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program. The computer program is operable to enable a computer to execute various processes of any of the above-described inspection path planning method embodiments and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0100] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0101] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0102] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for planning inspection paths, characterized in that, The method, applied to inspection robots, includes: During the inspection task movement based on the original path, the motion trajectory data of obstacles at multiple moments within a preset sliding time window are collected. The motion trajectory data includes the direction angle and velocity at each moment. Based on the collected motion trajectory data, the motion type of the obstacle is determined; Based on the movement type of the obstacle and the collected movement trajectory data, predict the movement path of the obstacle within a preset time period; The original path is adjusted based on the predicted motion path.

2. The method according to claim 1, characterized in that, Based on the collected motion trajectory data, the motion type of the obstacle is determined, including: Based on the direction angles at multiple moments in the collected motion trajectory data, the standard deviation of the direction angles is calculated. Based on the velocity at multiple moments in the collected motion trajectory data, the mean velocity, standard deviation of velocity, and mean acceleration are calculated respectively. The motion type of the obstacle is determined based on the standard deviation of the orientation angle, the mean velocity, the standard deviation of the velocity, and the mean acceleration.

3. The method according to claim 2, characterized in that, Based on the standard deviation of the orientation angle, the mean velocity, the standard deviation of the velocity, and the mean acceleration, the motion type of the obstacle is determined, including: If the standard deviation of the orientation angle is less than a first threshold, the standard deviation of the velocity is less than a second threshold, and the mean acceleration is less than a third threshold, then the motion type of the obstacle is determined to be uniform motion; or If the standard deviation of the orientation angle is less than the first threshold, the standard deviation of the velocity is less than the second threshold, and the mean acceleration is not less than the third threshold, then the motion type of the obstacle is determined to be a regular motion of accelerated motion; or If the standard deviation of the orientation angle is not less than the first threshold or the standard deviation of the velocity is not less than the second threshold, then the motion type of the obstacle is determined to be irregular motion; or If the direction angle and velocity at the multiple moments are both 0, then the motion type of the obstacle is determined to be stationary.

4. The method according to claim 1, characterized in that, Based on the obstacle's motion type and the collected motion trajectory data, predict the obstacle's motion path within a preset time period, including: If the obstacle's motion type is uniform motion, then the mean value of the direction angle is calculated based on the direction angles at multiple moments in the collected motion trajectory data; Calculate the average speed based on the speed at multiple moments in the collected motion trajectory data; Calculate the first moving distance corresponding to the preset duration based on the average speed; The movement path of the obstacle within a preset time period is predicted based on the mean azimuth angle and the first moving distance.

5. The method according to claim 1, characterized in that, Based on the obstacle's motion type and the collected motion trajectory data, predict the obstacle's motion path within a preset time period, including: If the obstacle's motion type is accelerated motion, then the mean value of the direction angle is calculated based on the direction angles at multiple moments in the collected motion trajectory data; Based on the velocity at multiple moments in the collected motion trajectory data, the mean velocity and mean acceleration are calculated respectively; The second moving distance corresponding to the preset duration is calculated based on the average velocity and the average acceleration. Based on the mean azimuth angle and the second moving distance, the movement path of the obstacle within a preset time period is predicted.

6. The method according to claim 1, characterized in that, Based on the obstacle's motion type and the collected motion trajectory data, predict the obstacle's motion path within a preset time period, including: If the obstacle's motion type is irregular, then the median velocity is calculated based on the velocity at multiple moments in the collected motion trajectory data; Calculate the third travel distance corresponding to the preset duration based on the median speed; The second obstacle avoidance length is obtained by multiplying the third moving distance and the preset safety factor. A circle is drawn with the center of the plane where the obstacle is currently located as the center and the second obstacle avoidance length as the radius to obtain the obstacle avoidance area; The obstacle avoidance area is predicted as the movement path of the obstacle within a preset time period.

7. The method according to claim 1, characterized in that, Based on the obstacle's motion type and the collected motion trajectory data, predict the obstacle's motion path within a preset time period, including: If the obstacle's motion type is stationary, then obtain the first intersection point where the obstacle's current position plane intersects with the original path; The first obstacle avoidance length is determined based on the length of the obstacle and the length of the inspection robot; With the first intersection point as the center and the third obstacle avoidance length as the radius, a circle is drawn to obtain the obstacle avoidance area; The obstacle avoidance area is predicted as the movement path of the obstacle within a preset time period.

8. The method according to claim 4 or 5, characterized in that, Based on the predicted motion path, the original path is adjusted, including: Determine the second intersection point between the predicted motion path and the original path; The first obstacle avoidance length is determined based on the length of the obstacle and the length of the inspection robot; With the second intersection point as the center and the first obstacle avoidance length as the radius, draw a circle to obtain the obstacle avoidance area; Obtain the third and fourth intersection points where the obstacle avoidance area intersects with the original path; The original path between the third and fourth intersection points is replaced with the outer edge of the first obstacle avoidance area between the third and fourth intersection points to obtain the adjusted path.

9. The method according to claim 6 or 7, characterized in that, Based on the predicted motion path, the original path is adjusted, including: Obtain the fifth and sixth intersection points where the obstacle avoidance area intersects with the original path; The original path between the fifth and sixth intersection points is replaced with the outer edge of the obstacle avoidance area between the fifth and sixth intersection points to obtain the adjusted path.

10. An inspection robot, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the method as described in any one of claims 1-9.

Images