Complex scene environmental sanitation robot obstacle avoidance method, device, equipment and medium

Abstract

The invention provides a complex scene environmental sanitation robot obstacle avoidance method and device, equipment and a medium. The method comprises the following steps: separating noise data from environment data to obtain a multi-source preliminary signal; space coordinates and velocity vectors of the dynamic obstacles are extracted, the positions of the dynamic obstacles, the motion trend of the target and the positions of the static obstacles are determined, and an obstacle avoidance sequence of the environmental sanitation robot is determined; extracting a key trigger point of obstacle avoidance and a feature vector thereof; judging whether the environmental sanitation robot has a deadlock potential area or not; if the deadlock potential area exists, adjusting the path weight to obtain an optimized planning path; adjusting parameters are extracted from the optimized planning path, and the relative displacement trend is determined; and performing priority ranking on the obstacles, determining a target obstacle avoidance path, and controlling the sanitation robot to move. Therefore, by separating the noise data from the environment data and formulating the collaborative obstacle avoidance strategy of the dynamic and static obstacles, the accuracy of obstacle avoidance in a complex scene can be improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving control, and particularly to an obstacle avoidance method, device, equipment and medium for a sanitation robot in a complex scenario. Background Art

[0002] The operation scenarios of sanitation robots are complex. In operation scenarios such as squares and scenic spots, the flow of people is large, there are static trash cans, road piles, and dynamic pedestrians, vehicles, etc. These obstacles all affect the actual cleaning operation effect of the cleaning robot. When facing complex scenarios, traditional path planning algorithms may get stuck in deadlocks due to being surrounded by obstacles or closed channels, or result in ineffective detours due to repeated path adjustments, reducing the cleaning efficiency. Therefore, the above-mentioned defects existing in the current sanitation robots during operation obstacle avoidance control are technical problems that need to be solved urgently. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide an obstacle avoidance method for a sanitation robot in a complex scenario, so as to improve the accuracy of obstacle avoidance in a complex scenario.

[0004] In a first aspect, the embodiments of the present application provide an obstacle avoidance method for a sanitation robot in a complex scenario, and the method includes: Separate noise data from the environmental data within the preset range of the sanitation robot to obtain multi-source preliminary signals, where the environmental data at least includes the speed and distance of dynamic obstacles and the distance of static obstacles within the preset range; Extract the spatial coordinates and velocity vectors of each dynamic obstacle from the multi-source preliminary signals, and based on the spatial mapping algorithm, determine the positions and target movement trends of each dynamic obstacle, and the positions of each static obstacle; Determine the obstacle avoidance sequence of the sanitation robot according to the positions and target movement trends of each dynamic obstacle and the positions of each static obstacle; Extract the key trigger points for the sanitation robot to avoid obstacles and the feature vectors of the key trigger points from the obstacle avoidance sequence; Based on the A* algorithm, determine whether there is a potential deadlock area for the sanitation robot according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot; In response to the sanitation robot having a potential deadlock area, adjust the path weight according to the potential deadlock area to obtain an optimized planned path, where the path weight at least includes a dynamic obstacle weight and a deadlock risk weight; Extract adjustment parameters from the optimized planned path to determine the relative displacement trends of each dynamic obstacle and each static obstacle, where the adjustment parameters include the speed of the sanitation robot; According to the relative displacement trend, prioritize each obstacle and determine the target obstacle avoidance path; Control the movement of the sanitation robot based on the target obstacle avoidance path.

[0005] In some embodiments, separate the noise data from the environmental data within the preset range of the sanitation robot to obtain multi-source preliminary signals, including: Divide the environmental data based on multiple independent signal channels to obtain multiple sub-channel data; Filter the multiple sub-channel data through a parallel filter to obtain the filtered signal; Separate the noise data from the filtered signal through an adaptive weighted average algorithm to obtain multi-source preliminary signals.

[0006] In some embodiments, extract the spatial coordinates and velocity vectors of each dynamic obstacle from the multi-source preliminary signals, and based on the spatial mapping algorithm, determine the positions and target movement trends of each dynamic obstacle, as well as the positions of each static obstacle, including: Decompose the multi-source preliminary signals to obtain the spatial coordinates and velocity vectors of each dynamic obstacle; Through the grid division method, construct a mapping structure based on the spatial coordinates and velocity vectors of each dynamic obstacle to obtain the initial spatial distribution data; Extract the positions corresponding to each static obstacle and the positions of each dynamic obstacle from the initial spatial distribution data; Determine the target movement trend of each dynamic obstacle according to the velocity vector.

[0007] In some embodiments, determining the target movement trend of each dynamic obstacle according to the velocity vector includes: Determine the initial movement trend of each dynamic obstacle according to the velocity vector; Based on the initial movement trends of each dynamic obstacle, determine the overlapping regions between each dynamic obstacle and each static obstacle; For the mapping structure within the overlapping region, separate each static obstacle and each dynamic obstacle by adjusting the grid to obtain an optimized obstacle point distribution, where the obstacle points include each static obstacle and each dynamic obstacle; According to the optimized obstacle point distribution, fuse the velocity vector to determine the target movement trend of each dynamic obstacle point; Wherein, the optimized obstacle point distribution is:

[0008] Wherein, represents the obstacle distribution density function, represents the number of sampling points, represents the weight coefficient, represents the attenuation coefficient, represents the position of the reference point, represents the position of the obstacle point.

[0009] In some embodiments, according to the positions of the dynamic obstacles and the target movement trend, as well as the positions of the static obstacles, an obstacle avoidance sequence for the sanitation robot is determined, including: Based on the time window sliding method, the overlapping areas of the dynamic obstacles are determined according to the positions of the dynamic obstacles and the target movement trend; When the overlapping area contains static obstacles, the priority index of each dynamic obstacle is determined by the vector decomposition method; Based on the priority indexes of the dynamic obstacles, the dynamic obstacles are sorted to determine the obstacle avoidance sequence of the sanitation robot; Wherein, the priority index is:

[0010] Wherein, represents the priority index, represents the priority coefficient, represents the weight factor, represents the obstacle distance, represents the scale parameter, represents the number of obstacles.

[0011] In some embodiments, adjustment parameters are extracted from the optimized planned path to determine the relative displacement trends of the dynamic obstacles and the static obstacles, including: Adjustment parameters are extracted from the optimized planned path, and the adjustment parameters include the speed of the sanitation robot; According to the distribution changes of the dynamic obstacles and the static obstacles, the environmental data is fused by a Kalman filter to obtain the preliminary displacement distribution of the dynamic obstacles; According to the preliminary displacement distribution, the overlapping areas of the dynamic obstacles and the static obstacles are extracted to determine the boundary ranges of the dynamic obstacles; According to the boundary ranges, the path planning parameters are updated to obtain an optimized path sequence; According to the optimized path sequence, the relative displacement trends of the dynamic obstacles and the static obstacles are determined.

[0012] In some embodiments, according to the relative displacement trend, performing priority sorting on each obstacle and determining a target obstacle avoidance path, including: Performing filtering and smoothing processing on the relative displacement trend to obtain a stable displacement trend distribution; Extracting the speeds of the dynamic obstacles according to the stable displacement trend distribution; Marking the dynamic obstacles with speeds exceeding a preset speed threshold as high-priority obstacles; For each of the high-priority obstacles, determining the absolute difference between the speed and the preset speed threshold; Sorting the absolute differences corresponding to the high-priority obstacles in descending order to obtain a target sorting sequence; Based on the target sorting sequence, determining the path deflection angle adjustment values for the dynamic obstacles; Determining a target obstacle avoidance path according to the path deflection angle adjustment values.

[0013] In a second aspect, an obstacle avoidance device for a sanitation robot in a complex scenario provided by an embodiment of the present application includes: A separation module, configured to separate noise data from environmental data within a preset range of the sanitation robot to obtain a multi-source preliminary signal, where the environmental data at least includes the speeds and distances of dynamic obstacles within the preset range and the distances of static obstacles; A first extraction module, configured to extract the spatial coordinates and velocity vectors of the dynamic obstacles from the multi-source preliminary signal, and based on a spatial mapping algorithm, determine the positions and target motion trends of the dynamic obstacles and the positions of the static obstacles; A first determination module, configured to determine an obstacle avoidance sequence of the sanitation robot according to the positions and target motion trends of the dynamic obstacles and the positions of the static obstacles; A second extraction module, configured to extract key trigger points for the sanitation robot to avoid obstacles and feature vectors of the key trigger points from the obstacle avoidance sequence; A judgment module, configured to judge whether there is a potential deadlock area for the sanitation robot based on the A* algorithm, according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot; An adjustment module, configured to, in response to the sanitation robot having a potential deadlock area, adjust the path weights according to the potential deadlock area to obtain an optimized planned path, where the path weights at least include a dynamic obstacle weight and a deadlock risk weight; A second determination module, configured to extract adjustment parameters from the optimized planned path and determine the relative displacement trends of the dynamic obstacles and the static obstacles, where the adjustment parameters include the speed of the sanitation robot; A third determination module, configured to perform priority sorting on each obstacle according to the relative displacement trend, and determine a target obstacle avoidance path; A control module, configured to control the movement of the sanitation robot based on the target obstacle avoidance path.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, including: A memory configured to store instructions; A processor configured to call the instructions from the memory and, when executing the instructions, be capable of implementing the complex scenario sanitation robot obstacle avoidance method provided in the first aspect of the embodiments of the present application.

[0015] In a fourth aspect, an embodiment of the present application provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to cause a machine to execute the complex scenario sanitation robot obstacle avoidance method according to the above.

[0016] In the embodiment of the present application, noise data is separated from the environmental data within the preset range of the sanitation robot to obtain multi-source preliminary signals. Among them, the environmental data at least includes the speed, distance of dynamic obstacles within the preset range, and the distance of static obstacles; the spatial coordinates and velocity vectors of each dynamic obstacle are extracted from the multi-source preliminary signals. Based on the spatial mapping algorithm, the positions and target movement trends of each dynamic obstacle, and the positions of each static obstacle are determined; according to the positions and target movement trends of each dynamic obstacle, and the positions of each static obstacle, the obstacle avoidance sequence of the sanitation robot is determined; the key trigger points for the sanitation robot to avoid obstacles and the feature vectors of the key trigger points are extracted from the obstacle avoidance sequence; based on the A* algorithm, according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot, it is judged whether there is a potential deadlock area for the sanitation robot; in response to the sanitation robot having a potential deadlock area, the path weight is adjusted according to the potential deadlock area to obtain an optimized planned path, where the path weight at least includes a dynamic obstacle weight and a deadlock risk weight; adjustment parameters are extracted from the optimized planned path to determine the relative displacement trends of each dynamic obstacle and each static obstacle, where the adjustment parameters include the speed of the sanitation robot; according to the relative displacement trends, priority sorting is performed on each obstacle to determine a target obstacle avoidance path; the movement of the sanitation robot is controlled based on the target obstacle avoidance path. In this way, by separating noise data from environmental data and formulating a cooperative obstacle avoidance strategy for dynamic and static obstacles, the accuracy of obstacle avoidance in complex scenarios can be improved. Description of the Drawings

[0017] Figure 1 is a schematic flowchart of the complex scenario sanitation robot obstacle avoidance method provided by the embodiment of the present application; Figure 2It is a trajectory diagram of obstacle avoidance for a sanitation robot in a complex scenario provided by a specific embodiment of the present application; Figure 3 It is a schematic structural diagram of an obstacle avoidance device for a sanitation robot in a complex scenario provided by an embodiment of the present application; Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Specific embodiments

[0018] Next, the technical solutions in the embodiments of the present application will be clearly described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0019] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same type, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0020] Next, in conjunction with the accompanying drawings, the obstacle avoidance method, device, equipment and medium for a sanitation robot in a complex scenario provided by the embodiments of the present application will be described in detail through specific embodiments and their application scenarios.

[0021] Please refer to Figure 1 , which is a schematic flowchart of an obstacle avoidance method for a sanitation robot in a complex scenario provided by an embodiment of the present application. This method is applied to an electronic device. As Figure 1 shown, the obstacle avoidance method for the sanitation robot in the complex scenario includes the following steps S100 to step S900.

[0022] Step S100: Separate the noise data from the environmental data within the preset range of the sanitation robot to obtain multi-source preliminary signals. Among them, the environmental data at least includes the speed and distance of dynamic obstacles and the distance of static obstacles within the preset range.

[0023] In the embodiments of the present application, the sanitation robot may include, but is not limited to, unmanned cleaning robots, manned cleaning robots, etc. The preset range of the sanitation robot can be understood as the data acquisition range of the current position of the sanitation robot set in advance. For example, it can be within 5 meters from the sanitation robot. The environmental data can be understood as multi-source data collected through sensors such as lidar and ultrasonic sensors. The sanitation data may include, but is not limited to, data such as the speed and distance of dynamic obstacles within the preset range, and the distance of static obstacles. Dynamic obstacles can be understood as obstacles that can move, such as pedestrians and vehicles. Static obstacles can be understood as obstacles that remain stationary, such as road studs, etc.

[0024] After collecting the environmental data within the preset range of the sanitation robot, the environmental data within the preset range of the sanitation robot can be filtered by the adaptive weighted average algorithm to separate the noise data, so as to obtain the multi-source preliminary signal after noise reduction.

[0025] Step S200: Extract the spatial coordinates and velocity vectors of each of the dynamic obstacles from the multi-source preliminary signal, and based on the spatial mapping algorithm, determine the positions and target movement trends of each of the dynamic obstacles, and the positions of each of the static obstacles.

[0026] In the embodiments of the present application, the spatial coordinates and velocity vectors can be obtained through the decomposition of the multi-source preliminary signal. In one example, by analyzing the frequency change of the multi-source preliminary signal, the relative velocity of the dynamic obstacle is calculated. According to the change of the position of the dynamic obstacle over time, the velocity vector of the dynamic obstacle can be estimated. In another example, the spatial coordinates of the dynamic obstacle can be estimated by using geometric positioning or the Doppler effect through information such as the intensity, time difference or frequency shift of the multi-source preliminary signal.

[0027] After obtaining the spatial coordinates and velocity vectors of each dynamic obstacle, a mapping structure can be constructed through the spatial mapping algorithm based on grid division to obtain the preliminary spatial distribution data. And extract the positions corresponding to each static obstacle and the positions of each dynamic obstacle from the initial spatial distribution data. Determine the target movement trend of each dynamic obstacle according to the velocity vector of each dynamic obstacle.

[0028] Step S300: Determine the obstacle avoidance sequence of the sanitation robot according to the positions and target movement trends of each of the dynamic obstacles, and the positions of each of the static obstacles.

[0029] In the embodiments of the present application, after determining the positions and target motion trends of the dynamic obstacles, based on the time window sliding method, the overlapping regions of the dynamic obstacles can be determined according to the positions and target motion trends of the dynamic obstacles. In the case where static obstacles are included in the overlapping regions, the priority indexes of the dynamic obstacles can be determined by the vector decomposition method. Then, based on the priority indexes of the dynamic obstacles, the dynamic obstacles are sorted, and the obstacle avoidance sequence of the sanitation robot can be determined.

[0030] Step S400: Extract the key trigger points for the sanitation robot to avoid obstacles and the feature vectors of the key trigger points from the obstacle avoidance sequence.

[0031] In the embodiments of the present application, the obstacle avoidance trigger points of the sanitation robot are extracted from the obstacle avoidance sequence, and a preset lightweight convolutional network is called to extract event features to obtain an initial feature set. Among them, the time features may include, but are not limited to, time features and spatial features (coordinates), etc. Then, the initial feature set is compressed by a sparse matrix to reduce the calculation delay and determine a refined feature set. For the trigger points in the refined feature set, it is judged whether the resource limit exceeds a preset threshold in the embedded environment to obtain a constraint condition. Among them, the resource limit may include, but is not limited to, device processor performance and memory, etc. If the threshold is exceeded, the sparse matrix structure is adjusted by a compression technique to obtain an optimized feature set. According to the correspondence between the optimized feature set and the trigger points, a distribution sequence of feature vectors is generated. The key trigger points are extracted from the distribution sequence, and the output of the lightweight convolutional network is fused to determine the feature vectors of the key trigger points.

[0032] Step S500: Based on the A* algorithm, determine whether there is a potential deadlock area for the sanitation robot according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot.

[0033] In the embodiments of the present application, the feature vectors of the key trigger points are combined with the preset operation duration of the sanitation robot, and the A* algorithm is used to evaluate the path deadlock risk to obtain a risk distribution. Then, the potential deadlock area of the sanitation robot is extracted from the risk distribution.

[0034] Step S600: In response to the sanitation robot having a potential deadlock area, adjust the path weight according to the potential deadlock area to obtain an optimized planned path, where the path weight includes at least a dynamic obstacle weight and a deadlock risk weight.

[0035] In the embodiments of the present application, when there is a potential deadlock area for the sanitation robot, the boundary of the potential deadlock area is detected to obtain a boundary division. According to the boundary division, the dynamic adjustment technology is used to update the path weights to obtain a weight distribution. The change trend of the local path is judged through the weight distribution, and an adjusted path set is obtained. For the key trigger points in the path set, combined with the complex parallel feature fusion strategy, an optimized path sequence can be determined. Among them, the path weights may include but are not limited to path cost (distance), dynamic obstacle weight, deadlock risk weight, etc., and the initial path sequence is finally adjusted and determined through different weight parameters. Then, the key area under time constraint is extracted from the initial path sequence, and a preset threshold is used for judgment to obtain an optimized planned path.

[0036] Step S700: Extract adjustment parameters from the optimized planned path, and determine the relative displacement trend between each dynamic obstacle and each static obstacle, where the adjustment parameters include the speed of the sanitation robot.

[0037] In the embodiments of the present application, adjustment parameters are extracted from the optimized planned path, where the adjustment parameters include the speed of the sanitation robot, and the speed of the sanitation robot may include the angular velocity of the sanitation robot.

[0038] For the changes in dynamic obstacle interference and static obstacle distribution, the Kalman filter is used to fuse multi-source data to obtain a preliminary displacement distribution of dynamic obstacles. The overlapping influence area of dynamic obstacles and static obstacles is extracted according to the preliminary displacement distribution. For the relative positions of dynamic obstacles and static obstacles, a preset threshold is used for judgment to determine the boundary range of dynamic interference.

[0039] The path planning parameters are updated through the boundary range of dynamic obstacles. For the change trend of the overlapping influence, an adjusted path weight distribution is obtained. The key nodes of the path are extracted according to the path weight distribution, and combined with the real-time adjustment information in the multi-source data, an optimized path sequence is obtained. The movement trend of dynamic obstacles is judged through the optimized path sequence. For the static constraints of the static obstacle distribution, the Kalman filter is used to update the relative displacement trend between each dynamic obstacle and each static obstacle.

[0040] In the embodiments of the present application, the intensity distribution of dynamic obstacles can be extracted according to the relative displacement trend, and combined with the adjustment parameters of the path, to obtain a final path optimization scheme.

[0041] Step S800: Perform priority sorting on each obstacle according to the relative displacement trend to determine a target obstacle avoidance path; In the embodiment of the present application, after obtaining the relative displacement trends of each dynamic obstacle and each static obstacle, a filtering method is used for smoothing processing to obtain a stable displacement trend distribution. Dynamic obstacle features are extracted according to the stable displacement trend distribution. If the speed exceeds a preset speed threshold, the dynamic obstacle is marked as a high-priority target. In one example, if the speed of a dynamic obstacle exceeds 1 m / s, the dynamic obstacle is marked as a high-priority target. An adjustment requirement range is determined according to each high-priority target. By combining the adjustment requirement range with a preset threshold, the priorities of each obstacle are sorted to obtain a target obstacle avoidance sequence, and then a target obstacle avoidance path can be determined.

[0042] In the embodiment of the present application, after determining the target obstacle avoidance sequence, dynamic obstacles with excessive speeds can be extracted according to the target obstacle avoidance sequence, parameters are adjusted to calculate the path deflection angle, and a deflection angle adjustment value is obtained. Among them, the deflection angle adjustment value can be expressed as:

[0043] Among them, represents the path deflection angle adjustment value, represents the deflection angle adjustment coefficient, and represent the target position coordinates, and represent the current position coordinates, represents the speed correction coefficient, represents the target speed, represents the speed threshold.

[0044] In this embodiment, the path planning is updated through the deflection angle adjustment value to generate a decision instruction sequence, and a preliminary obstacle avoidance plan is obtained. According to the preliminary obstacle avoidance plan, trend data is fused, and a Kalman filter is used to optimize and adjust parameters to judge the final decision instruction. By matching the dynamic target position with the final decision instruction, a real-time path adjustment sequence is obtained, and the target obstacle avoidance path is determined.

[0045] Step S900: Control the movement of the sanitation robot based on the target obstacle avoidance path.

[0046] Through the above steps S100 - S900, in the embodiments of the present application, noise data is separated from the environmental data within the preset range of the sanitation robot to obtain multi-source preliminary signals. Among them, the environmental data at least includes the speed and distance of dynamic obstacles within the preset range, and the distance of static obstacles; the spatial coordinates and velocity vectors of each dynamic obstacle are extracted from the multi-source preliminary signals, and based on the spatial mapping algorithm, the positions and target motion trends of each dynamic obstacle, as well as the positions of each static obstacle, are determined; according to the positions and target motion trends of each dynamic obstacle, and the positions of each static obstacle, an obstacle avoidance sequence of the sanitation robot is determined; the key trigger points for the sanitation robot to avoid obstacles and the feature vectors of the key trigger points are extracted from the obstacle avoidance sequence; based on the A* algorithm, according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot, it is judged whether there is a potential deadlock area for the sanitation robot; in response to the sanitation robot having a potential deadlock area, the path weight is adjusted according to the potential deadlock area to obtain an optimized planned path, where the path weight at least includes a dynamic obstacle weight and a deadlock risk weight; adjustment parameters are extracted from the optimized planned path to determine the relative displacement trends of each dynamic obstacle and each static obstacle, where the adjustment parameters include the speed of the sanitation robot; according to the relative displacement trends, the priorities of each obstacle are sorted to determine the target obstacle avoidance path; the movement of the sanitation robot is controlled based on the target obstacle avoidance path. In this way, by separating noise data from environmental data and formulating a cooperative obstacle avoidance strategy for dynamic and static obstacles, the accuracy of obstacle avoidance in complex scenarios can be improved.

[0047] In some embodiments, separating noise data from the environmental data within the preset range of the sanitation robot to obtain multi-source preliminary signals includes: Dividing the environmental data based on multiple independent signal channels to obtain multiple sub-channel data; Filtering the multiple sub-channel data through a parallel filter to obtain filtered signals; Separating noise data from the filtered signals through an adaptive weighted average algorithm to obtain multi-source preliminary signals.

[0048] Specifically, environmental data can be divided through multiple channels to form multiple independent signal channels, and multiple sub-channel data can be obtained. Among them, multiple channels can be understood as one acquisition device corresponding to one channel. Then, the multiple sub-channel data are processed by a parallel filter, and preset filtering parameters are used to separate high-frequency interference to obtain a filtered signal. For example, the displacement change of 0.5 m / s during the cleaning operation is tracked in real time. Then, the noise data are separated from the filtered signal through an adaptive weighted average algorithm. Specifically, for the filtered signal, the displacement change of the dynamic obstacle at adjacent time points is calculated, and the displacement change trend is determined through time series analysis. The adaptive weighted average algorithm is used to process the displacement change trend, and the weight is dynamically adjusted according to historical data to obtain a smooth displacement signal. If there are still abnormal fluctuations in the smooth displacement signal, the noise residue is judged through local mean detection to obtain an optimized signal. The multi-channel data are fused according to the optimized signal to obtain a multi-source preliminary signal after noise reduction.

[0049] In the embodiment of the present application, the displacement tracking result under low-speed cruise can be determined according to the multi-source preliminary signal. And by updating the multi-source preliminary signal in real time and combining the real-time acquired environmental data to adjust the filtering parameters, a continuous tracking signal of the dynamic obstacle displacement is obtained.

[0050] In some embodiments, the spatial coordinates and velocity vectors of each dynamic obstacle are extracted from the multi-source preliminary signal, and based on the spatial mapping algorithm, the positions and target motion trends of each dynamic obstacle, as well as the positions of each static obstacle are determined, including: Decompose the multi-source preliminary signal to obtain the spatial coordinates and velocity vectors of each dynamic obstacle; Through the grid division method, based on the spatial coordinates and velocity vectors of each dynamic obstacle, a mapping structure is constructed to obtain initial spatial distribution data; Extract the positions corresponding to each static obstacle and the positions of each dynamic obstacle from the initial spatial distribution data; Determine the target motion trend of each dynamic obstacle according to the velocity vector.

[0051] Specifically, through the decomposition of the multi-source preliminary signal, the spatial coordinates and velocity vectors of each dynamic obstacle can be obtained. In one example, by analyzing the frequency change of the multi-source preliminary signal, the relative velocity of the dynamic obstacle is calculated. According to the change of the position of the dynamic obstacle over time, the velocity vector of the dynamic obstacle can be estimated. In another example, the spatial coordinates of the dynamic obstacle can be estimated by using geometric positioning or Doppler effect through information such as the intensity, time difference or frequency offset of the multi-source preliminary signal.

[0052] Specifically, based on the grid division method, the spatial region is divided into grid cells of equal or unequal sizes. Data analysis or calculation is performed within each grid cell, and finally, the mapping of the entire spatial region is achieved. Each grid cell may contain data from different sensors or results calculated according to certain rules. Through reasonable grid division, effective analysis and processing of the space can be carried out.

[0053] After obtaining the spatial coordinates and velocity vectors of each dynamic obstacle, a spatial mapping algorithm based on grid division can be used to construct a mapping structure and obtain preliminary spatial distribution data. The position distribution corresponding to the static obstacles can be extracted from the initial spatial distribution data. And the target motion trends of each dynamic obstacle can be analyzed through the velocity vectors of each dynamic obstacle.

[0054] In some embodiments, determining the target motion trends of each of the dynamic obstacles according to the velocity vector includes: Determining the initial motion trends of each of the dynamic obstacles according to the velocity vector; Determining the overlapping regions between each of the dynamic obstacles and each of the static obstacles based on the initial motion trends of each of the dynamic obstacles; For the mapping structure within the overlapping region, by adjusting the grid, separating each of the static obstacles and each of the dynamic obstacles, an optimized distribution of obstacle points is obtained, where the obstacle points include each of the static obstacles and each of the dynamic obstacles; According to the optimized distribution of obstacle points, fusing the velocity vector to determine the target motion trends of each of the dynamic obstacle points; Wherein, the optimized distribution of obstacle points is:

[0055] Wherein, represents the obstacle distribution density function, represents the number of sampling points, represents the weight coefficient, represents the attenuation coefficient, represents the reference point position, represents the obstacle point position.

[0056] Specifically, the initial motion trends of each dynamic obstacle can be determined through the velocity vectors of each dynamic obstacle. Based on the initial motion trends of each dynamic obstacle and the positions of each static obstacle, the overlapping regions between each dynamic obstacle and each static obstacle can be determined. For the mapping structure within the overlapping region, a method of local grid adjustment can be adopted to separate the static obstacles and the dynamic obstacles, and an optimized distribution of obstacle points, that is, the distribution of static obstacles and dynamic obstacles, is obtained.

[0057] Specifically, according to the optimized distribution of obstacle points and by integrating the velocity vectors of dynamic obstacles, the target motion trends of each dynamic obstacle can be calculated. Among them, the velocity vector of a dynamic obstacle can be expressed as:

[0058] Among them, represents the velocity vector of the obstacle, represents the spatial position vector, , , respectively represent the velocity components in the x, y, and z directions.

[0059] The real-time position coordinates of a dynamic obstacle can be expressed as:

[0060] Among them, represents the real-time position coordinates of the obstacle, , and represent the initial coordinates, represents the coordinate correction amount.

[0061] Specifically, the optimized distribution of obstacle points can be detected through an obstacle detection function. The obstacle detection function can be expressed as:

[0062] Among them, represents the obstacle detection function, represents the detection threshold, represents the detection weight, represents the number of detection points, represents the spatial mapping function, represents the detection point coordinates.

[0063] In this embodiment, according to the target motion trends of each dynamic obstacle, the dynamic change directions of each dynamic obstacle can be determined. If the dynamic change direction exceeds the preset threshold, the grid division parameters are adjusted by comparing historical data to obtain an updated mapping structure. Then, a sequence of spatial coordinate changes of the dynamic obstacle is extracted from the updated mapping structure, and a weighted fusion method is used to integrate multi-source signals to determine the final motion trend distribution. By updating the velocity vector and spatial coordinates with the multi-source signals collected in real time and combining the adjusted mapping structure, the position distribution of the continuously tracked obstacle points can be obtained.

[0064] In some embodiments, according to the positions and target motion trends of each of the dynamic obstacles and the positions of each of the static obstacles, an obstacle avoidance sequence of the sanitation robot is determined, including: Based on the time window sliding method, determine the overlapping areas of the dynamic obstacles according to the positions and target motion trends of the dynamic obstacles; When static obstacles are included in the overlapping areas, determine the priority indexes of the dynamic obstacles by the vector decomposition method; Based on the priority indexes of the dynamic obstacles, sort the dynamic obstacles to determine the obstacle avoidance sequence of the sanitation robot; Among them, the priority index is:

[0065] Among them, represents the priority index, represents the priority coefficient, represents the weight factor, represents the obstacle distance, represents the scale parameter, represents the number of obstacles.

[0066] Specifically, after generating the positions and target motion trends of the dynamic obstacles by the grid division method, based on the time window sliding method, determine the overlapping areas of the dynamic obstacles according to the positions and target motion trends of the dynamic obstacles. And judge whether there are static obstacles in the overlapping areas. When static obstacles are included in the overlapping areas, the priority indexes of the dynamic obstacles can be determined by the vector decomposition method. Then sort the priority indexes of the dynamic obstacles in descending order to determine the obstacle avoidance sequence of the sanitation robot.

[0067] In this embodiment, the adjustment direction of the motion trend can be determined according to the obstacle avoidance sequence of the sanitation robot. For the adjusted motion trend, update the grid division structure to obtain optimized position data. Through the optimized position data and the obstacle avoidance sequence, fuse the time window sliding method to obtain the corrected trajectory of the dynamic obstacle. According to the corrected trajectory and the distribution of the static obstacles, judge the change of the motion trend after the separation of the dynamic obstacles, and then the obstacle avoidance sequence of the sanitation robot can be updated.

[0068] In some embodiments, extract adjustment parameters from the optimized planned path to determine the relative displacement trends of the dynamic obstacles and the static obstacles, including: Extract adjustment parameters from the optimized planned path, and the adjustment parameters include the speed of the sanitation robot; According to the distribution changes of the dynamic obstacles and the static obstacles, fuse the environmental data by using a Kalman filter to obtain the preliminary displacement distributions of the dynamic obstacles; Extract the overlapping regions of the dynamic obstacles and the static obstacles according to the preliminary displacement distribution, and determine the boundary ranges of the dynamic obstacles; Update the path planning parameters according to the boundary ranges to obtain an optimized path sequence; Determine the relative displacement trends of the dynamic obstacles and the static obstacles according to the optimized path sequence.

[0069] Specifically, in the embodiments of the present application, adjustment parameters are extracted from the optimized planned path, where the adjustment parameters include the speed of the sanitation robot. In view of the interference of dynamic obstacles and the change of the distribution of static obstacles, a Kalman filter is used to fuse multi-source environmental data to obtain the preliminary displacement distribution of the dynamic obstacles. The overlapping influence regions of the dynamic obstacles and the static obstacles are extracted according to the preliminary displacement distribution. For the relative positions of the dynamic obstacles and the static obstacles, a preset threshold is used for judgment to determine the boundary ranges of the dynamic interference.

[0070] Update the path planning parameters through the boundary ranges of the dynamic obstacles, and obtain the adjusted path weight distribution for the change trend of the overlapping influence. Extract the key nodes of the path according to the path weight distribution, and combine the real-time adjustment information in the multi-source data to obtain an optimized path sequence. Judge the movement trend of the dynamic obstacles through the optimized path sequence, and update the relative displacement trends of the dynamic obstacles and the static obstacles by using a Kalman filter for the static constraints of the static obstacle distribution.

[0071] In some embodiments, according to the relative displacement trends, priority sorting is performed on each obstacle to determine the target obstacle avoidance path, including: Perform filtering and smoothing processing on the relative displacement trends to obtain a stable displacement trend distribution; Extract the speeds of the dynamic obstacles according to the stable displacement trend distribution; Mark the dynamic obstacles with speeds exceeding the preset speed threshold as high-priority obstacles; For each of the high-priority obstacles, determine the absolute difference between the speed and the preset speed threshold; Sort the absolute differences corresponding to the high-priority obstacles in descending order to obtain a target sorting sequence; Based on the target sorting sequence, determine the path deflection angle adjustment values of the dynamic obstacles; Determine the target obstacle avoidance path according to the path deflection angle adjustment values.

[0072] Specifically, in the embodiments of the present application, after obtaining the relative displacement trends of each dynamic obstacle and each static obstacle, a filtering method is used for smoothing processing to obtain a stable displacement trend distribution. Dynamic obstacle features are extracted according to the stable displacement trend distribution. If the speed exceeds a preset speed threshold, the dynamic obstacle is marked as a high-priority target. In one example, if the speed of a dynamic obstacle exceeds 1 m / s, the dynamic obstacle is marked as a high-priority target. The adjustment requirement range is determined according to each high-priority target. By combining the adjustment requirement range with a preset threshold, the priority of each obstacle is sorted to obtain a target obstacle avoidance sequence, and then the target obstacle avoidance path can be determined.

[0073] In the embodiments of the present application, after determining the target obstacle avoidance sequence, the dynamic obstacles with excessive speed can be extracted according to the target obstacle avoidance sequence, the parameters are adjusted to calculate the path deflection angle, and the deflection angle adjustment value is obtained. Among them, the target speed can be expressed as:

[0074] Among them, represents the target speed, M represents the number of observed data points, represents the current position, represents the position at the previous moment, represents the time interval. This formula can be used to calculate the average speed of a dynamic obstacle; The deflection angle adjustment value can be expressed as:

[0075] Among them, represents the path deflection angle adjustment value, represents the deflection angle adjustment coefficient, and represent the target position coordinates, and represent the current position coordinates, represents the speed correction coefficient, represents the target speed, represents the speed threshold.

[0076] In this embodiment, the path planning is updated through the deflection angle adjustment value, a decision instruction sequence is generated, and a preliminary obstacle avoidance plan is obtained. According to the preliminary obstacle avoidance plan, the trend data is fused, and a Kalman filter is used to optimize the adjustment parameters to judge the final decision instruction. By matching the dynamic target position with the final decision instruction, a real-time path adjustment sequence is obtained, and the target obstacle avoidance path is determined.

[0077] Please refer to Figure 2 which is the trajectory diagram of obstacle avoidance of a sanitation robot in a complex scenario provided by a specific embodiment of the present application. As shown in Figure 2As shown in the figure, squares in the figure represent static obstacles, circles represent dynamic obstacles, and the line connecting the starting point to the ending point represents the obstacle avoidance path of the sanitation robot. The positions of the dynamic obstacles change in real time, and the sanitation robot adjusts the path deflection angle to achieve obstacle avoidance.

[0078] Please refer to Figure 3 , which is a schematic structural diagram of an obstacle avoidance device for a sanitation robot in a complex scenario provided by an embodiment of the present application. The second aspect of the embodiment of the present application provides an obstacle avoidance device 30 for a sanitation robot in a complex scenario. The obstacle avoidance device 30 for a sanitation robot in a complex scenario includes: A separation module 31, configured to separate noise data from environmental data within a preset range of the sanitation robot to obtain multi-source preliminary signals. Among them, the environmental data at least includes the speed and distance of dynamic obstacles within the preset range, and the distance of static obstacles. A first extraction module 32, configured to extract the spatial coordinates and velocity vectors of each dynamic obstacle from the multi-source preliminary signals, and based on a spatial mapping algorithm, determine the positions and target motion trends of each dynamic obstacle, and the positions of each static obstacle. A first determination module 33, configured to determine an obstacle avoidance sequence of the sanitation robot according to the positions and target motion trends of each dynamic obstacle, and the positions of each static obstacle. A second extraction module 34, configured to extract key trigger points for the sanitation robot to avoid obstacles from the obstacle avoidance sequence, and feature vectors of the key trigger points. A judgment module 35, configured to judge whether there is a potential deadlock area for the sanitation robot based on the A* algorithm, according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot. An adjustment module 36, configured to respond to the existence of a potential deadlock area for the sanitation robot, and adjust the path weight according to the potential deadlock area to obtain an optimized planned path, where the path weight at least includes a dynamic obstacle weight and a deadlock risk weight. A second determination module 37, configured to extract adjustment parameters from the optimized planned path, and determine the relative displacement trends of each dynamic obstacle and each static obstacle, where the adjustment parameters include the speed of the sanitation robot. A third determination module 38, configured to sort each obstacle according to the relative displacement trend to determine a target obstacle avoidance path. A control module 39, configured to control the movement of the sanitation robot based on the target obstacle avoidance path.

[0079] The obstacle avoidance device 30 for a sanitation robot in a complex scenario provided by the second aspect of the embodiment of the present application can implement each process implemented by the above method embodiment and achieve the same beneficial effects. To avoid repetition, it will not be elaborated here.

[0080] Please refer to Figure 4 , which is a schematic structural diagram of an electronic device provided by an embodiment of the present application. In a third aspect of the embodiments of the present application, an electronic device 4000 is provided, including a processor 4100 and a memory 4200. The memory 4200 stores machine-executable instructions that can be executed by the processor 4100. The processor 4100 can execute the machine-executable instructions to implement the above-mentioned obstacle avoidance method for a sanitation robot in a complex scenario.

[0081] In a fourth aspect of the embodiments of the present application, a machine-readable storage medium is provided. Instructions are stored on the machine-readable storage medium, and when the instructions are executed by a processor, the processor implements the above-mentioned obstacle avoidance method for a sanitation robot in a complex scenario.

[0082] In some embodiments, the embodiments of the present application also provide a computer program product, including a computer program, and the computer program implements the obstacle avoidance method for a sanitation robot in a complex scenario according to the above embodiments when executed by a processor.

[0083] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0084] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1the functions specified in one or more boxes. These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide for implementing in the process Figure 1 one process or more processes and / or boxes Figure 1 the steps of the functions specified in one box or more boxes.

[0085] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0086] The memory may include non-permanent memory in the computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

[0087] Computer-readable media includes permanent and non-permanent, removable and non-removable media and can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0088] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0089] The above are only the embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

[0090] In addition, any combination can be made between various different embodiments of the present invention, as long as it does not violate the idea of the present invention, and it should also be regarded as the content disclosed by the present invention.

Claims

1. A method for obstacle avoidance of a sanitation robot in a complex scenario, characterized in that, The method includes: Separating noise data from environmental data within a preset range of the sanitation robot to obtain multi-source preliminary signals, where the environmental data at least includes the speed and distance of dynamic obstacles and the distance of static obstacles within the preset range; Extracting the spatial coordinates and velocity vectors of each dynamic obstacle from the multi-source preliminary signals, and based on a spatial mapping algorithm, determining the positions and target motion trends of each dynamic obstacle, and the positions of each static obstacle; Determining an obstacle avoidance sequence for the sanitation robot according to the positions and target motion trends of each dynamic obstacle and the positions of each static obstacle; Extracting the key trigger points for the sanitation robot to avoid obstacles and the feature vectors of the key trigger points from the obstacle avoidance sequence; Based on the A-star algorithm, judging whether there is a potential deadlock area for the sanitation robot according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot; In response to the sanitation robot having a potential deadlock area, adjusting the path weight according to the potential deadlock area to obtain an optimized planned path, where the path weight at least includes a dynamic obstacle weight and a deadlock risk weight; Extracting adjustment parameters from the optimized planned path and determining the relative displacement trends of each dynamic obstacle and each static obstacle, where the adjustment parameters include the speed of the sanitation robot; Prioritizing each obstacle according to the relative displacement trends to determine a target obstacle avoidance path; Controlling the movement of the sanitation robot based on the target obstacle avoidance path.

2. The method according to claim 1, wherein The separating noise data from environmental data within a preset range of the sanitation robot to obtain multi-source preliminary signals includes: Dividing the environmental data based on multiple independent signal channels to obtain multiple sub-channel data; Performing filtering processing on the multiple sub-channel data through a parallel filter to obtain filtered signals; Separating noise data from the filtered signals through an adaptive weighted average algorithm to obtain multi-source preliminary signals.

3. The method according to claim 1, wherein The extracting the spatial coordinates and velocity vectors of each dynamic obstacle from the multi-source preliminary signals and, based on a spatial mapping algorithm, determining the positions and target motion trends of each dynamic obstacle and the positions of each static obstacle includes: Decomposing the multi-source preliminary signals to obtain the spatial coordinates and velocity vectors of each dynamic obstacle; Constructing a mapping structure based on the spatial coordinates and velocity vectors of each dynamic obstacle through a grid division method to obtain initial spatial distribution data; Extracting the positions corresponding to each static obstacle and the positions of each dynamic obstacle from the initial spatial distribution data; Determining the target motion trend of each dynamic obstacle according to the velocity vector.

4. The method according to claim 3, wherein The determining the target motion trend of each dynamic obstacle according to the velocity vector includes: Determining the initial motion trend of each dynamic obstacle according to the velocity vector; Determining the overlapping area of each dynamic obstacle and each static obstacle based on the initial motion trend of each dynamic obstacle; For the mapping structure within the overlapping region, by adjusting the grid, separate each of the static obstacles and each of the dynamic obstacles to obtain an optimized distribution of obstacle points, where the obstacle points include each of the static obstacles and each of the dynamic obstacles; Based on the optimized distribution of obstacle points, fuse the velocity vectors to determine the target motion trends of each of the dynamic obstacle points; Among them, the optimized distribution of obstacle points is: Among them, represents the obstacle distribution density function, represents the number of sampling points, represents the weight coefficient, represents the attenuation coefficient, represents the reference point position, represents the obstacle point position.

5. The method according to claim 1, characterized in that Determine the obstacle avoidance sequence of the sanitation robot according to the positions and target motion trends of each of the dynamic obstacles, and the positions of each of the static obstacles, including: Based on the time window sliding method, determine the overlapping region of each of the dynamic obstacles according to the positions and target motion trends of each of the dynamic obstacles; When the overlapping region contains static obstacles, determine the priority index of each dynamic obstacle by the vector decomposition method; Based on the priority index of each of the dynamic obstacles, sort each of the dynamic obstacles to determine the obstacle avoidance sequence of the sanitation robot; Among them, the priority index is: Among them, represents a priority index, represents a priority coefficient, represents a weight factor, represents the obstacle distance, represents a scale parameter, represents the number of obstacles.

6. The method according to claim 1, characterized in that, Extract adjustment parameters from the optimized planned path to determine the relative displacement trends of each of the dynamic obstacles and each of the static obstacles, including: Extract adjustment parameters from the optimized planned path, and the adjustment parameters include the speed of the sanitation robot; According to the distribution changes of each of the dynamic obstacles and static obstacles, use a Kalman filter to fuse the environmental data to obtain the preliminary displacement distribution of each of the dynamic obstacles; Extract the overlapping region of the dynamic obstacles and the static obstacles according to the preliminary displacement distribution to determine the boundary range of each of the dynamic obstacles; Update the path planning parameters according to the boundary range to obtain an optimized path sequence; Determine the relative displacement trends of each of the dynamic obstacles and each of the static obstacles according to the optimized path sequence.

7. The method according to claim 1, characterized in that According to the relative displacement trends, perform priority sorting on each obstacle to determine the target obstacle avoidance path, including: Perform filtering and smoothing processing on the relative displacement trends to obtain a stable displacement trend distribution; Extract the speeds of each of the dynamic obstacles according to the stable displacement trend distribution; Mark the dynamic obstacles with speeds exceeding the preset speed threshold as high-priority obstacles; For each of the high-priority obstacles, determine the absolute difference between the speed and the preset speed threshold; Sort the absolute differences corresponding to each of the high-priority obstacles in descending order to obtain a target sorting sequence; Based on the target sorting sequence, determine the path deflection adjustment value of each of the dynamic obstacles; Determine the target obstacle avoidance path according to the path deflection adjustment value.

8. An obstacle avoidance device for a sanitation robot in a complex scenario, characterized in that, The device includes: A separation module, configured to separate noise data from environmental data within a preset range of the sanitation robot to obtain multi-source preliminary signals, where the environmental data at least includes the speeds, distances of the dynamic obstacles within the preset range, and the distances of the static obstacles; The first extraction module is used to extract the spatial coordinates and velocity vectors of each dynamic obstacle from the multi-source preliminary signals, and based on the spatial mapping algorithm, determine the positions and target motion trends of each dynamic obstacle, as well as the positions of each static obstacle; The first determination module is used to determine the obstacle avoidance sequence of the sanitation robot according to the positions and target motion trends of each dynamic obstacle, and the positions of each static obstacle; The second extraction module is used to extract the key trigger points for the sanitation robot to avoid obstacles from the obstacle avoidance sequence, and the feature vectors of the key trigger points; The judgment module is used to judge whether there is a potential deadlock area for the sanitation robot based on the A* algorithm, according to the feature vectors of the key trigger points and the preset operation duration of the sanitation robot; The adjustment module is used to respond to the existence of a potential deadlock area for the sanitation robot, and adjust the path weights according to the potential deadlock area to obtain an optimized planned path, where the path weights at least include dynamic obstacle weights and deadlock risk weights; The second determination module is used to extract adjustment parameters from the optimized planned path and determine the relative displacement trends of each dynamic obstacle and each static obstacle, where the adjustment parameters include the speed of the sanitation robot; The third determination module is used to rank the priorities of each obstacle according to the relative displacement trends and determine the target obstacle avoidance path; The control module is used to control the movement of the sanitation robot based on the target obstacle avoidance path.

9. An electronic device, characterized in that, Comprising: A memory configured to store instructions; A processor configured to call the instructions from the memory and capable of implementing the complex scenario sanitation robot obstacle avoidance method according to any one of claims 1 to 7 when executing the instructions.

10. A machine-readable storage medium, characterized in that, Instructions are stored on the machine-readable storage medium, and the instructions are used to cause the machine to execute the complex scenario sanitation robot obstacle avoidance method according to any one of claims 1 to 7.

Images