Dynamic path planning method for large-scale grid map

Abstract

The invention relates to the technical field of path planning, and provides a dynamic path planning method for a large-scale grid map, which can construct three layers of strategy masks including a permission normal state layer, an execution state temporary extension layer and a service alarm layer so as to clearly divide path areas with different functions. Primary path planning is performed on a permission normal state layer based on a region-of-interest-thick line visual through straight pull search strategy, so that the conventional safety and stability of a primary path are ensured; detouring path planning is carried out on a permission normal state layer and an execution state temporary expansion layer based on a region of interest-thick line visual communication straight-pull search strategy, and real-time, efficient and safe ordered planning can be realized when a detouring demand is detected; according to the baseline path and the initial rendering path, regression convergence processing is carried out on the candidate detour path sub-segments, so that the safety and smoothness of the path are further improved; and the middle rendering path is subjected to angular point alignment resampling processing, so that visual illusion can be effectively avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning, in particular to a dynamic path planning method for large-scale grid maps. BACKGROUND

[0002] At present, for large-scale grid maps of areas such as warehouses, parks, factory areas, campuses and commercial complexes, robots usually need to rely on accurate path planning when performing tasks.

[0003] However, the path planning method in the prior art still has the following defects: (1) Real-time and safety are difficult to be compatible; (2) Business boundary is difficult to solidify: single mask is difficult to distinguish "initial planning" and "execution state bypass", and cannot finely manage temporary permissions and sensitive areas; (3) Display and execution are disconnected: front-end rendering linearization often causes visual illusion of "looking across obstacles"; (4) Lack of alarm management: areas that can pass through but are sensitive to business lack hierarchical response and traces; (5) Dynamic changes and accessibility: limited end-side computing power and frequent environmental dynamic changes make it difficult to achieve re-planning. SUMMARY

[0004] In view of the above, it is necessary to provide a dynamic path planning method, device, equipment and medium for large-scale grid maps, aiming to solve the problem of poor path planning effect of large-scale grid maps.

[0005] A dynamic path planning method for large-scale grid maps, the dynamic path planning method for large-scale grid maps comprising: In response to a robot path planning instruction based on an initial grid map, collecting environmental change data and processing the initial grid map according to the environmental change data to obtain a three-layer strategy mask, an integral graph and an off-obstacle cost graph; wherein the three-layer strategy mask includes a permission normal layer, an execution state temporary expansion layer and a business alarm layer; Based on a region of interest-thick line visual straight pull search strategy, performing initial path planning in the permission normal layer according to the integral graph and the off-obstacle cost graph to obtain a baseline path and an initial rendering path; Real-time monitoring of the execution state of the robot based on the baseline path and real-time environmental changes, and detecting whether a bypass event or an alarm event is triggered according to the execution state and the real-time environmental changes; When detecting that the bypass event is triggered, performing bypass path planning in the permission normal layer and the execution state temporary expansion layer based on the region of interest-thick line visual straight pull search strategy to obtain a candidate bypass path subsegment; converge the candidate detour path sub-segment according to the baseline path and the initial rendering path to obtain a target execution path and an intermediate rendering path; align and resample the intermediate rendering path to obtain a target rendering path; control the robot to execute a task based on the target execution path, and output the target rendering path to a designated display terminal.

[0006] A dynamic path planning device for a large-scale grid map, comprising: a processing unit configured to collect environmental change data in response to a robot path planning instruction based on an initial grid map, and process the initial grid map according to the environmental change data to obtain a three-layer strategy mask, an integral graph, and an off-obstacle cost graph; wherein the three-layer strategy mask comprises a permission normal layer, an execution state temporary expansion layer, and a business alarm layer; a planning unit configured to perform primary path planning on the permission normal layer according to the integral graph and the off-obstacle cost graph based on a region-of-interest-thick-line visual direct pull search strategy to obtain a baseline path and an initial rendering path; a detection unit configured to monitor an execution state of a robot based on the baseline path and real-time environmental changes in real time, and detect whether a detour event or an alarm event is triggered according to the execution state and the real-time environmental changes; the planning unit is further configured to perform detour path planning on the permission normal layer and the execution state temporary expansion layer based on the region-of-interest-thick-line visual direct pull search strategy to obtain a candidate detour path sub-segment when it is detected that the detour event is triggered; a converge unit configured to converge the candidate detour path sub-segment according to the baseline path and the initial rendering path to obtain a target execution path and an intermediate rendering path; a sampling unit configured to align and resample the intermediate rendering path to obtain a target rendering path; a control unit configured to control the robot to execute a task based on the target execution path, and output the target rendering path to a designated display terminal.

[0007] A computer device, comprising: a memory configured to store at least one instruction; and a processor configured to execute the instruction stored in the memory to implement the dynamic path planning method for a large-scale grid map.

[0008] A computer readable storage medium, the computer readable storage medium has at least one instruction stored therein, the at least one instruction is executed by a processor in a computer device to implement the dynamic path planning method of the large-scale grid map.

[0009] From the above technical solutions can be seen, the present application can construct three-layer policy mask including permission normal layer, execution state temporary expansion layer, service alarm layer, to clearly divide the path area of different functions; based on the thick line visual direct pull search strategy of the region of interest, the initial path planning is carried out in the permission normal layer, which guarantees the general safety and stability of the initial path; based on the thick line visual direct pull search strategy of the region of interest, the detour path planning is carried out in the permission normal layer and the execution state temporary expansion layer, which can realize real-time, efficient and safe orderly planning when the need for detour is detected; according to the baseline path and the initial rendering path, the candidate detour path subsegment is subjected to regression convergence processing, which further improves the safety and smoothness of the path; the intermediate rendering path is subjected to corner point alignment resampling processing, which can also effectively avoid visual artifacts. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 is a flow chart of a preferred embodiment of the dynamic path planning method of the large-scale grid map of the present application; Figure 2 is a functional module diagram of a preferred embodiment of the dynamic path planning device of the large-scale grid map of the present application; Figure 3 is a structural schematic diagram of a computer device for implementing the dynamic path planning method of the large-scale grid map of the present application. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be described in detail below with reference to the drawings and specific embodiments.

[0012] As Figure 1 shown, is a flow chart of a preferred embodiment of the dynamic path planning method of the large-scale grid map of the present application. The order of steps in the flow chart can be changed according to different needs, and some steps can be omitted.

[0013] The dynamic path planning method of the large-scale grid map is applied to one or more computer devices, which is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions. The hardware thereof includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0014] The computer device can be any electronic product that can interact with the user, such as a personal computer, a tablet computer, a smartphone, a personal digital assistant (PDA), a game console, an interactive Internet Protocol Television (IPTV), a smart wearable device, etc.

[0015] The computer device can also include network devices and / or user devices. The network devices include, but are not limited to, a single network server, a server group composed of multiple network servers, or a cloud composed of a large number of hosts or network servers based on cloud computing.

[0016] The server can be a standalone server or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, etc. Basic cloud computing services.

[0017] Among them, artificial intelligence (AI) is to use digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. Theory, method, technology and application system.

[0018] The basic technology of artificial intelligence generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. Artificial intelligence software technology mainly includes computer vision technology, robot technology, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc. Several major directions.

[0019] The network in which the computer device is located includes, but is not limited to, the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), and the like.

[0020] S10, in response to the robot path planning instruction based on the initial grid map, collecting environment change data, and processing the initial grid map according to the environment change data to obtain a three-layer strategy mask, a summed area table (SAT), and an obstacle distance cost map; wherein the three-layer strategy mask includes a permission normal layer, an execution state temporary expansion layer, and a business alarm layer.

[0021] In the embodiment, the initial grid map can be a large-scale grid map of a region such as a warehouse, a park, a factory, a campus, or a commercial complex.

[0022] In the embodiment, the robot can include a mobile robot such as a transfer robot.

[0023] In the embodiment, the environment change data can include, but is not limited to, obstacle position data, wall and other barrier data, and the like.

[0024] In the embodiment, the permission normal layer forms an obstacle distance buffer based on an inward bandwidth, and is used for inward morphological processing of a passable region; wherein the inward bandwidth is equal to the sum of the robot half-width and the positioning deviation range, and the inward bandwidth is dynamically adjusted according to the positioning accuracy; the cost model of the permission normal layer adopts a center preference-obstacle penalty strategy for path selection guidance, and does not contain a strategy penalty; The execution state temporary expansion layer is used for outward morphological processing of a temporary permission region; wherein the outward bandwidth is less than a configuration threshold; the execution state temporary expansion layer activation constraints include a time window constraint, a budget constraint, a no forbidden conflict constraint, and a concurrent capacity constraint; the strategy cost of the execution state temporary expansion layer includes an additional penalty for guiding preferential selection of the permission normal layer; the execution state temporary expansion layer revocation conditions include reaching the earliest return point, budget and time window depletion, manual revocation, and strategy withdrawal; the execution state temporary expansion layer includes a cumulative cost additional time growth item for encouraging the robot to return to the permission normal layer as soon as possible; The business alarm layer includes a region that is physically passable but is business sensitive or requires authorization; wherein when the robot enters the business alarm layer, a hierarchical alarm is triggered. The obstacle distance cost map is constructed based on a distance field, and is used for guiding the robot to preferentially select a path away from obstacles; wherein the distance field is calculated based on an Euclidean distance transform (EDT) algorithm.

[0025] Wherein, the positioning deviation range can be 3σ positioning, i.e. the inner shrink bandwidth = robot half-width + 3σ positioning, and σ can be configured according to actual needs.

[0026] Wherein, the off-obstacle buffer zone can reduce the cases of wall sticking and corner cutting.

[0027] Wherein, the center preference-off-obstacle penalty strategy is used to hope that the path of the robot is as close as possible to the center position of a passable area, and in order to ensure that the robot maintains a safe distance from the obstacle during movement and avoids collision, an additional penalty cost is applied to the path close to the obstacle.

[0028] Wherein, the inner shrink bandwidth can be dynamically adjusted according to the accuracy changes caused by scene switching such as day and night.

[0029] Wherein, the execution state temporary expansion layer is only enabled when the execution state local detour is performed, and does not participate in initial planning at ordinary times.

[0030] Wherein, the outer expansion bandwidth is less than the configured threshold, which can ensure that the expansion does not weaken the safety boundary.

[0031] Wherein, the time window constraint is used to limit within a time window; the budget constraint is used to limit the upper limit of distance, time, and number of times; the no forbidden conflict constraint is used to limit that the execution state temporary expansion layer and the business alarm layer, i.e. other forbidden rules, have no conflict; and the concurrent capacity constraint is used to limit the upper limit of the number of robots allowed to pass through the narrow passage and temporary detour area at the same time.

[0032] Wherein, the additional penalty is greater than 0, and the additional penalty can be adaptively adjusted according to type, priority, and time to guide the priority selection of the permission normal layer.

[0033] Wherein, the business alarm layer does not participate in planning by default.

[0034] In the business alarm layer, unauthorized is considered as forbidden, authorized or emergency allowed is considered as controlled passage, and entering means alarming.

[0035] Wherein, the integral graph can realize O(1) query of sub-window pixel sum, which is convenient for quickly counting the proportion of safe pixels in the area.

[0036] Of course, a distance field can also be used instead of the integral graph for subsequent calculation, which is not described here.

[0037] Through the above embodiment, different function path areas can be clearly divided, the normal state layer is allowed to guarantee normal safety passing, the execution state temporary expansion layer meets temporary bypassing requirements, the business alarm layer realizes sensitive area management and control, and clear strategy basis is provided for subsequent path planning.

[0038] In addition, the embodiment supports dynamic adjustment of parameters such as bandwidth in the normal state permission layer, so that the map preprocessing result can adapt to changes in different environments and positioning accuracy, and the flexibility and adaptability of planning are improved.

[0039] S11, based on a Region of Interest (ROI) -thick line visual direct pull search strategy, a baseline path and an initial rendering path are obtained by performing initial path planning on the normal state permission layer according to the integral graph and the obstacle distance cost graph.

[0040] In the embodiment, the initial path planning on the normal state permission layer according to the integral graph and the obstacle distance cost graph based on the Region of Interest-thick line visual direct pull search strategy includes: obstacle blocking data and positioning error are generated according to the obstacle distance cost graph with the normal state permission layer as a basic range constraint; obtain the queue situation in the three-layer strategy mask; generate a first region of interest according to the obstacle blocking data, the positioning error and the queue situation, and search for a plurality of potential paths in the first region of interest; sample along the plurality of potential paths according to a preset step length to obtain a plurality of sampling points; For the neighborhood window of each sampling point, a thick line visual check algorithm is used to calculate the safety pixel ratio based on the integral graph, and the sampling point is determined to be safe when the safety pixel ratio is greater than a preset threshold. obtain the potential path in which all sampling points are safe as a candidate path; perform path point safety simplification processing on the candidate path to obtain the baseline path and the initial rendering path.

[0041] The queue situation is used to reflect the queuing situation of the robot.

[0042] The preset step length can be customized, such as 0.5-1.0px.

[0043] The preset threshold value can also be customized, such as 0.80, 0.95.

[0044] Wherein, for the case that the neighborhood window part exceeds the range of the grid map (such as the sampling point at the edge of the map, half of the window is in the map and half is out of the map). At this time, if the "out-of-bound part of the grid" is directly counted, it will cause a safety judgment error due to "no data", so a special grid boundary processing rule is needed. Specifically, the effective pixel count can be normalized.

[0045] Wherein, for sub-pixels or jaggies, 2x supersampling can be performed to suppress false judgments of jaggies.

[0046] Wherein, if the distance field is used instead of the integral graph for initial path planning, when the sampling point safety detection is performed, the corresponding EDT value is directly queried after sampling to determine whether it is greater than the sum of the preset radius and the check value. When the EDT values of all sampling points are greater than the sum of the preset radius and the check value, it is determined to be safe.

[0047] Correspondingly, for the case that the neighborhood window part exceeds the range of the grid map, if the distance field is used, the out-of-bound is considered as 0 distance; for sub-pixels or jaggies, bilinear interpolation is enabled.

[0048] Wherein, if it is detected to be unsafe, the last safe sampling point is returned or segmented bisection is performed until the longest safe sub-section is found.

[0049] Wherein, the search algorithm of the potential path can be flexibly selected to improve applicability.

[0050] In the embodiment, the path point safety simplification processing on the candidate path includes: Obtaining corners less than an included angle threshold from the candidate path as true corners, and retaining the sampling points corresponding to the true corners as true corner points; Inhibiting simplification again within a preset angle after the corner; Merging the sampling points with collinear relationship in the candidate path; Identifying a spike line segment in the candidate path, and merging the spike line segment to a neighboring steady-state line segment; Eliminating close-fitting points of the start point and the end point in the candidate path; Wherein, all sampling points need to remain safe when performing path point safety simplification processing; Wherein, after performing path point safety simplification processing, the adjacent line segment corner and the rotatable radius of the underlying controller remain matched.

[0051] Wherein, the included angle threshold can be configured as 25°-35°. Retaining the sampling points corresponding to the true corners can force to retain substantial turns and ensure the safety of the robot turning.

[0052] Wherein, the simplification is prohibited again within a preset angle after the corner, which can prevent the continuous turning and ensure the smooth transition of the path.

[0053] Wherein, the sampling points with the collinear relationship in the candidate path are merged, which can effectively simplify the path.

[0054] Wherein, the spike line segment is merged into the adjacent stable line segment, which can remove the line segment with large turning amplitude and short arms, and avoid unnecessary turning and shaking of the robot.

[0055] Wherein, the close-fitting points at the starting point and the ending point of the candidate path are removed, which is beneficial to the starting of the robot and the convergence of the path.

[0056] Wherein, in order to match the turning radius of the adjacent line segment with the bottom controller, a short arc or a transition point can be inserted when necessary.

[0057] Through the above embodiments, only the normal state layer planning is allowed, which ensures the general safety and stability of the initial path and meets the basic needs of daily patrol. The thick line visual inspection algorithm ensures the safety of the path line segment level, avoiding safety hazards such as cutting corners, scraping edges, and passing through narrow gaps. The safe simplification of the path point simplifies the path without sacrificing safety, improving the efficiency and stability of the robot execution.

[0058] S12, real-time monitoring the execution state of the robot based on the baseline path and real-time environmental changes, and detecting whether to trigger a detour event or an alarm event according to the execution state and the real-time environmental changes.

[0059] In this embodiment, the detection of whether to trigger a detour event or an alarm event according to the execution state and the real-time environmental changes comprises: obtaining real-time positioning data of the robot according to the execution state, and detecting whether there is an obstacle blocking the current execution path according to the real-time positioning data, the real-time environmental changes, and the baseline path, to obtain a blocking detection result; detecting whether there is a queuing event and a yielding event according to the real-time positioning data and the real-time environmental changes, to obtain a queue-yielding detection result; calculating the deviation of the actual position of the robot from the baseline path according to the real-time positioning data and the baseline path, to obtain a positioning deviation detection result; real-time detection of task change information; predicting whether the alarm event will occur according to the execution state and the real-time environmental changes, to obtain an alarm event prediction result; According to the blocking detection result, the queue-yielding detection result, the positioning deviation detection result, and the task change information, it is determined whether to trigger the detour event, and according to the warning event prediction result, it is determined whether to trigger the warning event.

[0060] When the occupation probability rises, a new obstacle is found in static image difference, a car or a pedestrian cuts in, or the like is detected, it can be determined that there is an obstacle blocking the current execution path.

[0061] When the concurrent capacity limit hits, the queuing delay exceeds the threshold, or the like is detected, it is determined that there is a queuing event and a yielding event.

[0062] The positioning deviation detection result can include tracking error, angle with baseline path, lateral drift, and the like.

[0063] The task change information can include task priority switching, emergency notification, time window about to expire, and the like.

[0064] The warning event prediction result can be used to reflect whether the robot will enter the business warning layer area in the subsequent execution process, so as to make preparations in advance.

[0065] Through the above embodiments, various abnormal situations in the execution process can be found in time through real-time monitoring, and safety accidents or task failures caused by environmental changes or abnormal states of the robot can be avoided.

[0066] S13, when the detour event is detected, a detour path planning is performed in the permission normal layer and the execution state temporary expansion layer based on the region of interest-thick line visual direct pull search strategy, and a candidate detour path subsegment is obtained.

[0067] In this embodiment, the detour path planning based on the region of interest-thick line visual direct pull search strategy in the permission normal layer and the execution state temporary expansion layer to obtain the candidate detour path subsegment includes: Obtain the current pose of the robot, the obstacle envelope, and the time step; Predict the predicted position of the robot after passing the time step; Determine the minimum bounding box according to the current pose, the obstacle envelope, and the predicted position; Expand the minimum bounding box by a preset multiple to obtain a second region of interest; Detect whether the execution state temporary expansion layer meets the enable constraint; When the execution state temporary expansion layer meets the enable constraint, the union of the permission normal layer and the execution state temporary expansion layer in the second region of interest is determined as a passable region. Based on the policy cost, path search is performed in the passable area to obtain a plurality of candidate sub-segments; The thick line visual check algorithm is used for line segment level safety detection on the plurality of candidate sub-segments; The candidate sub-segments passing the line segment level safety detection are determined as the candidate detour path sub-segments.

[0068] The preset multiple can be configured according to actual needs, such as 1.2-1.5 times.

[0069] When detecting whether the execution state temporary expansion layer meets the enabling constraint, if the number of occupants of the current execution state temporary expansion layer area is greater than the upper limit of the execution state temporary expansion layer concurrent capacity, the robot detouring can be applied to join the queue, and the scheduling is performed according to the principle of "priority first -> same level first in first out -> timeout anti-starvation", and the short job is prioritized to break the tie, and the estimated entry time is periodically estimated, if the estimated entry time is greater than the service commitment threshold or the budget will exceed, the robot is guided to change the route or give up using the execution state temporary expansion layer.

[0070] When the path search is performed in the passable area, a search algorithm compatible with the initial planning can be used. In the search process, the additional penalty of the execution state temporary expansion layer needs to be considered to guide the path to pass through the execution state temporary expansion layer as little as possible.

[0071] Through the above embodiments, the local region of interest re-planning avoids global recalculation, combines with an efficient search algorithm, guarantees the real-time of the detour planning, meets the scene demand of large-scale grid map under the limited end-side computing power, the multi-dimensional constraint verification of the execution state temporary expansion layer enables ensures the compliance and safety of the detour, avoids the chaos and risk caused by disordered detour, the concurrent capacity arbitration mechanism takes into account the efficiency and fairness, reduces congestion and meeting deadlock, and the set range search and thick line visual check expand the possibility of the detour path while ensuring the safety of the detour path, and avoid introducing new safety hazards due to detour.

[0072] S14, performing regression convergence processing on the candidate detour path sub-segments according to the baseline path and the initial rendering path to obtain a target execution path and an intermediate rendering path.

[0073] In this embodiment, the regression convergence processing on the candidate detour path sub-segments according to the baseline path and the initial rendering path to obtain a target execution path and an intermediate rendering path includes: A plurality of candidate regression points are selected one by one along the potential splicing area of the candidate detour path sub-segment and the baseline path; The thick-line visual inspection algorithm is used to perform safety detection on a line segment from the candidate detour path subsegment to each candidate return point and a line segment from each candidate return point to a subsequent segment of the baseline path, to obtain a thick-line visual inspection safety verification result; The total path cost after splicing the plurality of candidate return points is calculated, and an initial path cost of a corresponding path in the baseline path is calculated. The total path cost and the initial path cost are compared to obtain a cost superiority verification result; The alignment error of the candidate detour path subsegment and the baseline path at each candidate return point is calculated to obtain a geometric alignment verification result; When it is detected that the thick-line visual inspection safety verification result corresponding to any candidate return point is that the line segment from the candidate detour path subsegment to each candidate return point and the line segment from each candidate return point to the subsequent segment of the baseline path all pass the safety detection, the cost superiority verification result corresponding thereto is that the total path cost is less than the difference between the initial path cost and a jitter value, and the geometric alignment verification result corresponding thereto is that the alignment error is less than an error threshold, it is determined that the any candidate return point meets a safety condition, and it is further detected whether a continuous preset number of candidate return points all meet the safety condition; When the continuous preset number of candidate return points all meet the safety condition, the any candidate return point is determined to be an effective return point; An earliest return point is selected from all effective return points as a point closest to a detour starting point; A part of the candidate detour path subsegment from a corresponding starting point to the earliest return point and a part from the earliest return point to a terminal point of the baseline path are spliced to obtain the target execution path; The execution-state temporary extension layer is closed, and execution data of the execution-state temporary extension layer is emptied; The target execution path is rendered to obtain the intermediate rendering path.

[0074] In the selecting of the plurality of candidate return points one by one along the potential splicing area of the candidate detour path subsegment and the baseline path, the candidate return point needs to be located at an intersection or adjacent area of the candidate detour path subsegment and the baseline path, and it is necessary to ensure the continuity of the path from the candidate detour path subsegment to the candidate return point and from the candidate return point to the subsequent segment of the baseline path.

[0075] The thick-line visual inspection algorithm can ensure that both line segments meet the safety requirements and have no problems such as no cutting angle and no edge wiping.

[0076] The jitter value is used for anti-shake to avoid misjudgment due to small cost fluctuations.

[0077] The cost superior verification can ensure that the path after regression is superior or not worse than the original path in cost.

[0078] The alignment error includes position deviation, angle deviation, etc.

[0079] The geometric alignment verification can ensure smooth transition of the spliced path, and avoid large steering or shaking of the robot.

[0080] The preset number can be 2-3. The effective regression point is determined only when the continuous 2-3 candidate regression points are safe, which can avoid misjudgment caused by instantaneous data fluctuation and ensure the stability of regression.

[0081] If the candidate regression point is still in the execution state temporary expansion layer region and the budget or time window is about to be exhausted, the regression point that is earlier and meets the budget or time window requirement needs to be selected first to ensure successful regression to the normal layer before the resources are exhausted.

[0082] The determination of the earliest regression point can achieve regression to the baseline path as soon as possible and reduce the residence time in the execution state temporary expansion layer.

[0083] The execution data can include time growth items, budget measurement accumulation, etc., and stop the strategy punishment and arbitration related to the execution state temporary expansion layer. After completing the path splicing and closing the execution state temporary expansion layer, the robot returns to the state of executing the baseline path and returns to the execution monitoring phase.

[0084] In the above embodiment, the multi-criterion fusion verification of the earliest regression point ensures the safety, cost superiority and smoothness of the regression path, avoids safety hazards or path mutations after regression, and reduces the residence time of the robot in the execution state temporary expansion layer. The execution state temporary expansion layer can reduce the cost increase caused by the additional punishment of the execution state temporary expansion layer, and avoid long-term occupation of the resources of the execution state temporary expansion layer, improve the use efficiency of the execution state temporary expansion layer, and provide resources for other robots that need to detour.

[0085] S15, performing angle point alignment resampling processing on the intermediate rendering path to obtain a target rendering path.

[0086] In this embodiment, the angle point alignment resampling processing on the intermediate rendering path to obtain a target rendering path includes: performing phase resetting on the intermediate rendering path to obtain a plurality of sampling segments; wherein, when performing phase resetting, all original polyline vertices are retained; For each sampling segment, the previous sampling point is determined as a first sampling point, and a next sampling point is calculated from the first sampling point along the sampling segment according to a sampling step size; it is detected whether the next sampling point enters a buffer zone at the end of the sampling segment; when the next sampling point enters the buffer zone, an end point of the sampling segment is determined as a target point; when the next sampling point does not enter the buffer zone, a safety detection is performed on a line segment formed by the first sampling point and the next sampling point by using the thick-line visual check algorithm, and when the safety detection is passed, the next sampling point is determined as the target point, and a next sampling point is continuously calculated until all sampling segments are processed; when the line segment formed by the first sampling point and the next sampling point does not pass the safety detection, an end point of the sampling segment is inserted as a bridging point into the sampling segment, the first sampling point is updated as the bridging point, and a next sampling segment formed by the bridging point is continuously processed. The target rendering path is generated according to the original polyline vertex, the target point and the bridging point.

[0087] All original polyline vertices are reserved, no over-smoothing processing is performed on a rendering side, corner features of the path are accurately presented, and semantic information such as a path inflection point, a deceleration point and a prompt point is convenient for a user to identify through a visual interface.

[0088] The target rendering path is used for monitoring.

[0089] Through the above embodiment, the segment-by-segment phase resetting and the forced conformal ensure that a display path can accurately reflect geometric features and semantic information of an original path, avoid path distortion caused by over-smoothing, and facilitate a user to intuitively understand an actual driving track and key nodes of a robot; the thick-line visual check bottoming mechanism completely eliminates a visual illusion that a display line penetrates through an obstacle, ensures that what is seen is what is done, and makes a path observed by a user through a visual interface and a path actually executed by the robot strictly consistent in safety, thereby improving trust of the user in path planning; the segment-end buffer zone setting ensures that a sampling point accurately falls at a vertex at an end of a line segment, avoids deviation of the display path at the vertex caused by sampling errors, and further improves accuracy and aesthetics of the display path.

[0090] S16, controlling the robot to execute a task based on the target execution path, and outputting the target rendering path to a designated display terminal.

[0091] The designated display terminal can be a robot control terminal or the like.

[0092] In the embodiment, after the path planning based on the region-of-interest-thick-line-visual-check direct search strategy is performed on the permission normal layer and the execution-state temporary expansion layer, the method further includes: When the detour path planning fails, the preset multiple is increased step by step to expand the second region of interest until the preset multiple reaches an upper limit, and the detour path planning based on the region-of-interest-thick-line visual straight pull search strategy is continued in the expanded second region of interest; and / or The strategy cost is adjusted in parameters; and / or The neighborhood window is adjusted; and / or The detour path planning record is obtained, and the detour path planning record is reported to a designated operation and maintenance terminal, and the robot is guided to a safe waiting position.

[0093] Wherein, the anisotropic amplification along the normal layer skeleton can be performed, and the expansion to the passable area is preferred, and the expansion to the meaningless obstacle area is avoided.

[0094] Wherein, the off-obstacle cost weight, the smoothness weight and the like can be adjusted by a certain proportion (such as 15%) to reduce the requirements for the path off-obstacle distance and the smoothness, so that the search algorithm is more likely to find a feasible path. At the same time, the size of the execution state temporary expansion layer penalty can be adjusted appropriately.

[0095] Wherein, the original domain restriction can be switched to a more relaxed neighborhood restriction, the moving direction selection of the robot is increased, and the path search space is expanded.

[0096] Wherein, the detour path planning record can include detailed information of a re-planning failure, such as a constraint condition relaxed at each stage, a search result, a map data snapshot, a strategy parameter and the like.

[0097] Wherein, the robot waits for a manual instruction or a strategy update at the safe waiting position; if the manually adjusted map data or the strategy parameter (such as a newly added execution state temporary expansion layer region or an adjusted budget) is received, the robot re-enters the detour triggering stage to attempt re-planning; and if a task termination instruction is received, the current patrol task is ended.

[0098] In the above embodiment, the constraints are relaxed in stages and in order (ROI→cost→neighborhood), which avoids a large decline in path safety caused by indiscriminate relaxation, maximizes the path accessibility under the premise of ensuring safety, and reduces task failures caused by local obstacles; thick line visual check is performed after each relaxation to ensure that the found path still meets the basic safety requirements even in the case of constraint relaxation, and to avoid introducing new safety hazards; the reporting after failure and the safe waiting mechanism ensure that the robot is in a safe state when it cannot find a path autonomously, and provide complete data support for manual intervention, which facilitates quick positioning and solving of problems, and ensures the overall continuity of the patrol task.

[0099] In the embodiment, after detecting whether to trigger the bypass event or the warning event according to the execution state and the real-time environment change, the method further comprises: When detecting the triggering of the warning event, performing hierarchical response according to the execution state and the real-time environment change, and monitoring and recording the execution data of the robot in the business warning layer in real time; When monitoring that the robot leaves the business warning layer, detecting whether the robot meets the return condition; When the robot meets the return condition, stopping the hierarchical response, and controlling the robot to restore the normal state in the baseline path.

[0100] For example, when the distance between the current position of the robot and the business warning layer region reaches a pre-warning threshold, or the robot is about to enter the business warning layer according to the task planning, the warning event pre-warning can be triggered. Alternatively, the authorization information of the robot is queried to determine whether the authorization to enter the business warning layer is obtained and the authorization level; if the authorization is not obtained or the authorization has expired, the response such as forced stop and detour is performed, and the event is immediately reported; if the authorization is obtained, the corresponding hierarchical response measure is determined according to the authorization level and the sensitive level of the business warning layer.

[0101] The hierarchical response can include: L1 level response (prompting): suitable for the case that the authorization level is low and the sensitive degree of the business warning layer is not high, prompting information is displayed through a user interaction interface, voice broadcast prompt, light flashing (such as yellow light), and the like, reminding the operator that the robot is about to enter or has entered the business warning layer region, and recording the response start time; L2 level response (speed limit / sounding): suitable for the case that the authorization level is medium and the sensitive degree of the business warning layer is high, in addition to performing the L1 level prompting measure, the upper limit of the speed of the robot is lowered to a preset value (such as 0.6 m / s, which can be configured according to the requirement of the business warning layer), and a sounding device (such as a buzzer) is started to warn the surrounding personnel; at the same time, the derived cost of the business warning layer region is raised in the cost function, guiding the robot to leave the business warning layer as soon as possible; recording the speed limit value, the sounding start time, and the like; L3 level response (forced stop / detour): suitable for the case that the authorization to enter the business warning layer is not obtained, the authorization is expired, or the sensitive degree of the business warning layer is extremely high (such as involving a core area or a dangerous area), immediately stopping the robot running (if the robot has entered the business warning layer), or guiding the robot to deviate from the original path to bypass the business warning layer region (if the robot is about to enter the business warning layer); at the same time, generating emergency warning information and reporting to the management system to notify the relevant personnel to handle; recording the forced stop position, the detour path, the warning time, and the like.

[0102] The execution data can include the position, speed, surrounding environment change and other states of the robot in the business alarm layer, to ensure that the response measures are effectively executed, and if abnormal conditions occur (such as speed exceeding the limit, deviating from the specified path), the response measures are adjusted in time (such as stopping again, enhancing the prompt).

[0103] The execution data can also include key information of the alarm event recorded in real time, such as the time of entering the business alarm layer, the time of leaving the business alarm layer, the identification of the business alarm layer, the penalty coefficient of the business alarm layer, the speed limit value, the authorization information, the execution of the response measures, etc., while generating a strategy signature based on the budget snapshot, time window and other parameters, to ensure that the information is auditable and traceable.

[0104] The position of the robot after leaving the business alarm layer can be detected to determine whether it is within a reasonable range of the baseline path, whether the speed has returned to normal, whether the surrounding environment meets the conditions for continuing to execute the original task, etc. If the regression conditions are met, the hierarchical response measures are stopped, and the normal speed and execution state of the robot are restored. At the same time, all recorded information of the alarm event can be archived and stored, the strategy signature is updated, the robot returns to the execution monitoring phase, and continues to execute the patrol task according to the baseline path.

[0105] In the above embodiments, the hierarchical response mechanism differentiates the treatment according to the sensitivity of the business alarm layer and the authorization situation, ensuring the safety and compliance of the business alarm layer area while avoiding excessive control affecting the task execution efficiency, achieving flexible and rigorous management of sensitive areas; full information recording and strategy signature archiving provide complete and traceable data support for subsequent audit of alarm events, meeting the needs of business compliance and safety audit, facilitating responsibility identification and problem troubleshooting; real-time state monitoring and abnormal adjustment ensure the effective execution of hierarchical response measures, avoiding safety accidents or violations of robots in the business alarm layer area, and ensuring the safety of the business alarm layer area and the safety of the robots themselves.

[0106] In this embodiment, relevant parameters and execution records of the entire path planning and execution process can also be recorded, and a signature and a comprehensive audit report can be generated to form a governance closed loop, facilitating subsequent tracing. Comprehensive budget, time window and concurrent capacity control ensure that the path planning and execution process meet the business resource limitations and time requirements, avoiding low efficiency or conflicts caused by resource waste and disordered use, and improving the refinement of business management; the alarm event audit and strategy signature mechanism ensures the compliance and data security of sensitive area management, realizes traceability and responsibility identification of events, and meets the needs of industry supervision and enterprise internal safety management; the comprehensive audit report provides intuitive and comprehensive business operation data for management personnel, facilitating the analysis of business bottlenecks, the optimization of strategy parameters (such as adjusting the budget, time window and concurrent capacity), the improvement of path planning algorithm, and the promotion of the continuous optimization and upgrading of the entire patrol system.

[0107] From the above technical solutions can be seen, the present application can construct three-layer policy mask including permission normal layer, execution state temporary expansion layer, business alarm layer, to clearly divide the path area of different functions; based on the interested area-thick line visual direct pull search strategy, the initial path planning is carried out in the permission normal layer, which guarantees the general safety and stability of the initial path; based on the interested area-thick line visual direct pull search strategy, the detour path planning is carried out in the permission normal layer and the execution state temporary expansion layer, which can realize real-time, efficient and safe orderly planning when the detour demand is detected; the regression convergence processing is carried out on the candidate detour path subsegment according to the baseline path and the initial rendering path, which further improves the safety and smoothness of the path; the corner point alignment resampling processing is carried out on the intermediate rendering path, which can also effectively avoid visual artifacts.

[0108] As Figure 2 shown is a functional module diagram of a preferred embodiment of the dynamic path planning device of the large-scale grid map of the present application. The dynamic path planning device 11 of the large-scale grid map comprises a processing unit 110, a planning unit 111, a detection unit 112, a convergence unit 113, a sampling unit 114 and a control unit 115. The module / unit referred to in the present application refers to a series of computer program segments capable of being executed by a processor and capable of completing a fixed function, which is stored in a memory. In the present embodiment, the functions of each module / unit will be described in detail in subsequent embodiments.

[0109] The processing unit 110 is used for collecting environmental change data in response to a robot path planning instruction based on an initial grid map, and processing the initial grid map according to the environmental change data to obtain a three-layer policy mask, an integral graph and an obstacle distance cost graph; wherein the three-layer policy mask comprises a permission normal layer, an execution state temporary expansion layer and a business alarm layer. The planning unit 111 is used for performing initial path planning in the permission normal layer based on the integral graph and the obstacle distance cost graph according to the interested area-thick line visual direct pull search strategy, to obtain a baseline path and an initial rendering path. The detection unit 112 is used for monitoring the execution state of the robot based on the baseline path and real-time environmental changes, and detecting whether a detour event or an alarm event is triggered according to the execution state and the real-time environmental changes. The planning unit 111 is also used for performing detour path planning in the permission normal layer and the execution state temporary expansion layer based on the interested area-thick line visual direct pull search strategy when it is detected that the detour event is triggered, to obtain a candidate detour path subsegment. The convergence unit 113 is configured to perform regression convergence processing on the candidate detour path sub-segment according to the baseline path and the initial rendering path, to obtain a target execution path and an intermediate rendering path. The sampling unit 114 is configured to perform corner point alignment resampling processing on the intermediate rendering path, to obtain a target rendering path. The control unit 115 is configured to control the robot to execute a task based on the target execution path, and output the target rendering path to a designated display terminal.

[0110] From the above technical solutions, it can be seen that the present application can construct a three-layer policy mask including a permission normal layer, an execution state temporary expansion layer and a service alarm layer, to clearly divide the path area of different functions; the initial path planning is performed on the permission normal layer based on the ROI-thick line visual direct pull search strategy, to ensure the general safety and stability of the initial path; the detour path planning is performed on the permission normal layer and the execution state temporary expansion layer based on the ROI-thick line visual direct pull search strategy, to realize real-time, efficient and safe orderly planning when the detour demand is detected; the regression convergence processing is performed on the candidate detour path sub-segment according to the baseline path and the initial rendering path, to further improve the safety and smoothness of the path; the corner point alignment resampling processing is performed on the intermediate rendering path, to effectively avoid visual artifacts.

[0111] As shown in Figure 3 FIG. 1 is a structural schematic diagram of a computer device according to the preferred embodiment of the present application.

[0112] The computer device 1 can include a memory 12, a processor 13 and a bus (the arrow in the figure is the bus), and can further include a computer program stored in the memory 12 and executable on the processor 13, such as a large-scale grid map dynamic path planning program.

[0113] Those skilled in the art can understand that the schematic diagram is only an example of the computer device 1, and does not constitute a limitation on the computer device 1, which can be a bus type structure or a star type structure, and can further include more or less other hardware or software, or different component arrangements, such as the computer device 1 can further include an input / output device, a network access device, etc.

[0114] It should be noted that the computer device 1 is only an example, and other existing or future electronic products, such as those adaptable to the present application, should also be included within the protection scope of the present application and included by reference.

[0115] The memory 12 includes at least one type of readable storage medium, such as a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 12 can be an internal storage unit of the computer device 1, such as a mobile hard disk of the computer device 1. In other embodiments, the memory 12 can also be an external storage device of the computer device 1, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 12 can include both an internal storage unit and an external storage device of the computer device 1. The memory 12 can be used to store application software and various data installed in the computer device 1, such as a code of a dynamic path planning program for a large-scale raster map, etc., and can also be used to temporarily store data that has been output or will be output.

[0116] The processor 13 can be composed of an integrated circuit in some embodiments, such as a single packaged integrated circuit or a plurality of packaged integrated circuits with the same or different functions, including one or more combinations of a central processing unit (CPU), a microprocessor, a digital processing chip, a graphics processor, and various control chips, etc. The processor 13 is a control unit of the computer device 1, and is connected to various components of the computer device 1 through various interfaces and lines, and executes various functions and processes data of the computer device 1 by running or executing programs or modules stored in the memory 12 (e.g., a dynamic path planning program for a large-scale raster map, etc.) and calling data stored in the memory 12.

[0117] The processor 13 executes an operating system and various application programs installed in the computer device 1. The processor 13 executes the application programs to implement the steps in each of the above-described dynamic path planning methods for a large-scale raster map, such as Figure 1 the steps shown in the above embodiments.

[0118] By way of example, the computer program can be divided into one or more modules / units, which are stored in the memory 12 and executed by the processor 13 to accomplish the present application. The one or more modules / units can be a series of computer-readable instruction segments capable of accomplishing a specific function, which are used to describe the execution process of the computer program in the computer device 1. For example, the computer program can be divided into a processing unit 110, a planning unit 111, a detection unit 112, a convergence unit 113, a sampling unit 114, and a control unit 115.

[0119] The integrated units implemented in the form of software function modules described above can be stored in a computer-readable storage medium. The software function modules described above are stored in a storage medium, including a number of instructions for causing a computer device (which can be a personal computer, a computer device, or a network device, etc.) or a processor to execute part of the large-scale grid map dynamic path planning method described in various embodiments of the present application.

[0120] The modules / units integrated in the computer device 1, if implemented in the form of software function modules and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the present application can also be implemented by a computer program to instruct related hardware devices to complete all or part of the processes in the above-described embodiments. The computer program can be stored in a computer-readable storage medium and can implement the steps of each method embodiment when executed by a processor.

[0121] The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer-readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory, etc.

[0122] Further, the computer-readable storage medium can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function, etc.; and the data storage area can store data created according to the use of the blockchain node, etc.

[0123] The blockchain referred to in the present application is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm. The blockchain is essentially a decentralized database, and is a series of data blocks associated using cryptographic methods. Each data block contains information about a batch of network transactions, and is used to verify the validity of the information (anti-fake) and generate the next block. The blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.

[0124] The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one straight line is used in the Figure 3 , but it does not mean that there is only one bus or one type of bus. The bus is arranged to enable connection and communication between the memory 12, the at least one processor 13, and the like.

[0125] Although not shown, the computer device 1 can also include a power supply (such as a battery) for powering the various components. Preferably, the power supply can be logically connected to the at least one processor 13 through a power management device, so that the power management device can realize functions such as charge management, discharge management, and power consumption management. The power supply can also include one or more direct current or alternating current power supplies, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and the like. The computer device 1 can also include various sensors, Bluetooth modules, Wi-Fi modules, and the like, which will not be described here.

[0126] Further, the computer device 1 can also include a network interface. Optionally, the network interface can include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is usually used to establish a communication connection between the computer device 1 and other computer devices.

[0127] Optionally, the computer device 1 can further comprise a user interface, which can be a display, an input unit such as a keyboard, and optionally a standard wired interface, a wireless interface. Optionally, in some embodiments, the display can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc. The display can also be appropriately referred to as a display screen or a display unit, for displaying information processed in the computer device 1 and for displaying a visualized user interface.

[0128] It should be understood that the embodiments are only for illustration and do not limit the scope of the patent application.

[0129] Those skilled in the art can understand that, Figure 3 The structure shown does not constitute a limitation on the computer device 1, which can include fewer or more components than shown, or combine certain components, or different component arrangements.

[0130] In combination Figure 1 The memory 12 in the computer device 1 stores a plurality of instructions to implement a dynamic path planning method for a large-scale grid map, and the processor 13 can execute the plurality of instructions to implement: In response to a robot path planning instruction based on an initial grid map, collect environmental change data and process the initial grid map according to the environmental change data to obtain a three-layer strategy mask, an integral graph, and an obstacle distance cost graph; wherein the three-layer strategy mask includes a permission normal layer, an execution state temporary expansion layer, and a business alarm layer; Based on a region-of-interest-thick-line visual direct pull search strategy, perform initial path planning on the permission normal layer according to the integral graph and the obstacle distance cost graph to obtain a baseline path and an initial rendering path; Real-time monitor the execution state of the robot based on the baseline path and real-time environmental changes, and detect whether a detour event or an alarm event is triggered according to the execution state and the real-time environmental changes; When it is detected that the detour event is triggered, perform detour path planning on the permission normal layer and the execution state temporary expansion layer based on the region-of-interest-thick-line visual direct pull search strategy to obtain a candidate detour path subsegment; Perform regression convergence processing on the candidate detour path subsegment according to the baseline path and the initial rendering path to obtain a target execution path and an intermediate rendering path; Perform corner alignment resampling processing on the intermediate rendering path to obtain a target rendering path; The robot is controlled to perform a task based on the target execution path, and the target rendering path is output to a designated display terminal.

[0131] Specifically, the processor 13 can refer to the specific implementation method of the above instructions Figure 1 The description of the related steps in the corresponding embodiments will not be repeated here.

[0132] It should be noted that the data involved in the present case are all legally obtained. The non-company software tools or components appearing in the embodiments of the present application are only examples for introduction and do not represent actual use.

[0133] In several embodiments provided by the present application, it should be understood that the disclosed system, device and method can be implemented by other ways. For example, the device embodiments described above are only schematic, and for example, the division of the modules is only a logical function division, and there can be other division manners in actual implementation.

[0134] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0135] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical units, i.e. they can be located in one place or distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment.

[0136] In addition, the functional modules in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of hardware plus software functional modules.

[0137] It will be obvious to a person skilled in the art that the application is not limited to the details of the above-described exemplary embodiments but can be implemented in other embodiments without departing from the scope of the application.

[0138] The embodiments should therefore be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the description given above, so that all changes coming within the meaning and equivalency range of the claims are intended to be embraced therein. Any reference signs in the claims should not be construed as limiting the scope of the claims.

[0139] Furthermore, the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural and vice-versa, unless the context clearly requires these exclusions. The composition of the application can be either hardware-only or software-only or a combination of hardware and software.

[0140] Finally, it should be noted that the above-mentioned embodiments illustrate rather than limit the application, since the scope of the application is to be determined by the appended claims.

Claims

1. A dynamic path planning method for large-scale raster maps, characterized in that, The dynamic path planning method for large-scale raster maps includes: In response to robot path planning instructions based on an initial grid map, environmental change data is collected, and the initial grid map is processed according to the environmental change data to obtain a three-layer policy mask, an integral map, and an obstacle departure cost map; wherein, the three-layer policy mask includes a permitted normal layer, an execution-state temporary extension layer, and a service alarm layer; Based on the region of interest-thick line view direct pull search strategy, the initial path planning is performed on the permitted normal layer according to the integral map and the obstacle removal cost map to obtain the baseline path and the initial rendering path. The robot's execution status based on the baseline path and real-time environmental changes are monitored in real time, and whether a detour event or alarm event is triggered based on the execution status and real-time environmental changes. When the detour event is detected, detour path planning is performed in the permitted normal layer and the execution temporary extension layer based on the region of interest-thick line view direct pull search strategy to obtain candidate detour path segments; Based on the baseline path and the initial rendering path, the candidate detour path segments are subjected to regression convergence processing to obtain the target execution path and intermediate rendering path. The intermediate rendering path is resampled with corner alignment to obtain the target rendering path; The robot is controlled to perform tasks based on the target execution path, and the target rendering path is output to the designated display terminal.

2. The dynamic path planning method for large-scale raster maps as described in claim 1, characterized in that: The permitted normal layer forms an obstacle avoidance buffer zone based on the shrinking bandwidth, which is used to perform shrinking morphological processing on the passable area; wherein, the shrinking bandwidth is equal to the sum of the robot's half-width and the positioning deviation range, and the shrinking bandwidth is dynamically adjusted with the positioning accuracy; the cost model of the permitted normal layer adopts a center preference-obstacle avoidance penalty strategy for path selection guidance, and does not contain strategy penalty. The execution-state temporary extension layer is used to perform outward morphological processing on the temporary permitted area; wherein, the outward bandwidth is less than the configured threshold; the activation constraints of the execution-state temporary extension layer include time window constraints, budget constraints, no prohibited conflict constraints, and concurrent capacity constraints; the policy cost of the execution-state temporary extension layer includes: additional penalties to guide the robot to prioritize the permitted normal layer; the revocation conditions of the execution-state temporary extension layer include: reaching the earliest return point, budget and time window exhaustion, manual revocation, and policy withdrawal; the execution-state temporary extension layer includes a cumulative cost plus time growth term to encourage the robot to return to the permitted normal layer as soon as possible; The business alarm layer includes physically accessible but business-sensitive or authorized areas; wherein, when the robot enters the business alarm layer, a tiered alarm is triggered. The obstacle departure cost map is constructed based on a distance field to guide the robot to prioritize paths away from obstacles; wherein, the distance field is calculated based on the Euclidean distance transformation algorithm.

3. The dynamic path planning method for large-scale raster maps as described in claim 2, characterized in that, The region-of-interest-thick-line view-through direct pull search strategy, based on the integral map and the obstacle departure cost map, performs initial path planning on the permitted normal layer to obtain the baseline path and initial rendering path, including: Using the permitted normal layer as the basic range constraint, obstacle blocking data and positioning error are generated based on the obstacle departure cost map; Obtain the queue status within the three-layer policy mask; A first region of interest is generated based on the obstacle blocking data, the positioning error, and the queue situation, and multiple potential paths are searched within the first region of interest. Sampling is performed along the multiple potential paths according to a preset step size to obtain multiple sampling points; For the neighborhood window of each sampling point, the thick line view verification algorithm is used to calculate the safe pixel ratio based on the integral image, and the sampling point is determined to be safe when the safe pixel ratio is greater than a preset threshold. Obtain potential paths that are safe at all sampling points as candidate paths; The candidate path is subjected to path point safety simplification to obtain the baseline path and the initial rendering path.

4. The dynamic path planning method for large-scale raster maps as described in claim 3, characterized in that, The process of performing path point safety simplification on the candidate path includes: From the candidate paths, obtain the corners that are less than the included angle threshold as true corners, and retain the sampling points corresponding to the true corners as true corner points; Further simplification is prohibited within the preset angle after the corner; Merge the sample points that are collinear in the candidate paths; Identify spike segments in the candidate paths and merge the spike segments into adjacent steady-state segments; Eliminate the closest points to the start and end points in the candidate paths; Among them, all sampling points must remain safe when performing path point safety simplification processing; In particular, after performing path point safety simplification, the turning angle of adjacent line segments is matched with the rotatable radius of the underlying controller.

5. The dynamic path planning method for large-scale raster maps as described in claim 1, characterized in that, The step of detecting whether a detour event or alarm event is triggered based on the execution state and the real-time environmental changes includes: The robot's real-time positioning data is obtained based on the execution status, and based on the real-time positioning data, the real-time environmental changes, and the baseline path, it is detected whether there are obstacles blocking the current execution path, and the blocking detection result is obtained. Based on the real-time location data and the real-time environmental changes, the queue-yield detection result is obtained to determine whether there are queuing events and yielding events. The deviation between the robot's actual position and the baseline path is calculated based on the real-time positioning data and the baseline path to obtain the positioning deviation detection result; Real-time detection of task change information; Based on the execution status and the real-time environmental changes, predict whether the alarm event will occur, and obtain the alarm event prediction result; Whether to trigger the detour event is determined based on the blocking detection result, the queue-giveaway detection result, the positioning deviation detection result, and the task change information; and whether to trigger the alarm event is determined based on the alarm event prediction result.

6. The dynamic path planning method for large-scale raster maps as described in claim 3, characterized in that, The region-of-interest-thick-line-view-direct-pull search strategy performs detour path planning in the permitted normal layer and the execution-state temporary extension layer, resulting in candidate detour path segments including: Obtain the robot's current pose, obstacle envelope, and time step; Predict the robot's position after the time step; Determine the minimum bounded bounding box based on the current pose, the obstacle envelope, and the predicted position; The minimum bounded bounding box is expanded by a preset multiple to obtain the second region of interest; Detect whether the execution-state temporary extension layer satisfies the enable constraint; When the execution-state temporary extension layer satisfies the enable constraint, the union of the permitted normal layer and the execution-state temporary extension layer in the second region of interest is determined as a passable region. Based on the strategy cost, a path search is performed within the passable area to obtain multiple candidate sub-segments; The thick line visibility verification algorithm is used to perform line segment-level security detection on the multiple candidate sub-segments; Candidate sub-segments that pass the line segment-level safety detection are determined as candidate detour path sub-segments.

7. The dynamic path planning method for large-scale raster maps as described in claim 3, characterized in that, The step of performing regression convergence processing on the candidate detour path segments based on the baseline path and the initial rendering path to obtain the target execution path and intermediate rendering path includes: Multiple candidate regression points are selected one by one along the potential splicing area between the candidate bypass path segment and the baseline path; The thick line vision verification algorithm is used to perform security checks on the line segments from the candidate detour path sub-segments to each candidate regression point and the line segments from each candidate regression point to the subsequent segments of the baseline path, to obtain the thick line vision security verification results. Calculate the total path cost after concatenating the multiple candidate regression points, and calculate the initial path cost of the corresponding path in the baseline path; compare the total path cost with the initial path cost to obtain the cost superiority verification result; Calculate the alignment error between the candidate detour path segment and the baseline path at each candidate regression point to obtain the geometric alignment verification result; When the thick line visibility security verification result corresponding to any candidate regression point is that the line segments from the candidate detour path sub-segment to each candidate regression point and the line segments from each candidate regression point to the subsequent segments of the baseline path all pass the security detection, the corresponding cost superiority verification result is that the total path cost is less than the difference between the initial path cost and the jitter value, and the corresponding geometric alignment verification result is that the alignment error is less than the error threshold, it is determined that the arbitrary candidate regression point meets the security conditions, and the detection continues to check whether a consecutive preset number of candidate regression points all meet the security conditions; When the predetermined number of consecutive candidate regression points all meet the safety conditions, any candidate regression point is determined as a valid regression point. Select the point closest to the starting point of the detour from all valid regression points as the earliest regression point; The candidate detour path segment from the corresponding starting point to the earliest regression point is concatenated with the segment from the earliest regression point to the end point of the baseline path to obtain the target execution path; Close the execution-state temporary extension layer and clear the execution data of the execution-state temporary extension layer; The target execution path is rendered to obtain the intermediate rendering path.

8. The dynamic path planning method for large-scale raster maps as described in claim 3, characterized in that, The step of performing corner alignment resampling on the intermediate rendering path to obtain the target rendering path includes: The intermediate rendering path is phase-reset segment by segment to obtain multiple sampling segments; wherein, when performing phase-reset segment by segment, all original polyline vertices are retained; For each sampling segment, the previous sampling point is designated as the first sampling point, and the next sampling point is calculated along the sampling segment, starting from the first sampling point, according to the sampling step size. It is then checked whether the next sampling point enters the buffer at the end of the sampling segment. When the next sampling point enters the buffer, the end point of the sampling segment is designated as the target point. When the next sampling point does not enter the buffer, the thick line visibility check algorithm is used to perform a security check on the line segment formed by the first sampling point and the next sampling point. If the security check is passed, the next sampling point is designated as the target point, and the calculation of the next sampling point continues until all sampling segments are processed. Wherein, when the line segment formed by the first sampling point and the next sampling point fails the security check, the end point of the sampling segment is inserted as a bridging point into the sampling segment, the first sampling point is updated as the bridging point, and the processing of the next sampling segment formed by the bridging point continues. The target rendering path is generated based on the original polyline vertices, the target point, and the bridging point.

9. The dynamic path planning method for large-scale raster maps as described in claim 6, characterized in that, After the region of interest-thick line view direct pull search strategy performs detour path planning in the permitted normal layer and the execution state temporary extension layer, the method further includes: When detour path planning fails, the preset multiplier is incrementally increased to expand the second region of interest until the preset multiplier reaches its upper limit. Within the expanded second region of interest, detour path planning continues based on the region of interest-thick line view direct pull search strategy; and / or Adjust the parameters of the strategy cost; and / or Adjust the neighborhood window; and / or Obtain the detour path planning record, report the detour path planning record to the designated operation and maintenance terminal, and guide the robot to a safe waiting position.

10. The dynamic path planning method for large-scale raster maps as described in claim 3, characterized in that, After detecting whether a detour event or alarm event is triggered based on the execution state and the real-time environmental changes, the method further includes: When the alarm event is detected, a graded response is performed according to the execution status and the real-time environmental changes, and the robot's execution data in the business alarm layer is monitored and recorded in real time. When the robot is detected to have left the business alarm layer, it is checked whether the robot meets the regression conditions; When the robot meets the regression condition, it stops performing hierarchical responses and controls the robot to return to the normal state within the baseline path.

Images