Mobile body path planning method and system based on intelligent cabinet multi-mode sensing and lithium battery power supply

Abstract

The invention relates to the technical field of intelligent moving body path planning, in particular to a moving body path planning method and system based on intelligent cabinet multi-mode sensing and lithium battery power supply, and the method comprises the following steps: S1, a moving body obtains a target point coordinate, a task priority parameter and a current electric quantity state; s2, environment data are collected, and a three-dimensional grid map with confidence scores is generated; s3, calculating the maximum reachable radius of the moving body, and generating an initial candidate path set; s4, monitoring environment change data and an electric quantity attenuation rate in real time, and dynamically adjusting a path node weight; s5, when the remaining power cannot reach the next node, preferentially selecting the intelligent cabinet with the minimum deviation degree according to the path deviation degree; and S6, outputting the final path to a moving body control system. According to the invention, dynamic path optimization and intelligent energy supplement scheduling under the condition of limited energy consumption are realized by fusing multi-modal sensing, electric quantity evaluation and intelligent cabinet path cooperation mechanisms.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent mobile body path planning, and in particular to a mobile body path planning method and system based on multimodal perception of an intelligent cabinet and lithium battery power supply. Background Art

[0002] When mobile objects perform tasks such as intelligent inspections, terminal delivery, and environmental monitoring, the accuracy and stability of path planning are directly related to operational efficiency and system safety. Traditional path planning systems typically rely on a single sensor for environmental mapping, which is limited by perception accuracy and information redundancy, making it difficult to meet path planning requirements in complex dynamic environments. At the same time, with the widespread use of lithium batteries in mobile objects, path planning must not only consider spatial accessibility but also energy consumption constraints.

[0003] However, most existing solutions separate path planning from functional modules such as battery assessment and environmental perception, lacking an integrated intelligent path generation and dynamic adjustment mechanism. This is particularly true in scenarios with multiple concurrent tasks and non-fixed charging points. Mobile objects may run out of battery while performing tasks. Without timely access to the recharging path reconstruction mechanism, this can lead to task interruptions or even system failures. Therefore, a mobile object path planning method and system based on multimodal sensing in smart cabinets and lithium battery power is urgently needed to address these issues. Summary of the Invention

[0004] Based on the above objectives, the present invention provides a mobile object path planning method and system based on multimodal perception of smart cabinets and lithium battery power supply.

[0005] The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply includes the following steps: S1: The mobile object obtains the target point coordinates and task priority parameters, and simultaneously uploads the current power status of the lithium battery to the nearest smart cabinet; S2: The smart cabinet collects environmental data through a multimodal sensor array and generates a three-dimensional grid map with confidence scores; S3: Based on the current power state of S1, the three-dimensional grid map of S2, and the lithium battery discharge model, the maximum reachable radius of the mobile body is calculated to generate an initial set of candidate paths; S4: Real-time monitoring of environmental change data and battery attenuation rate of the moving object, and dynamic adjustment of path node weights; S5: When the remaining power cannot reach the next node, the smart cabinet location database is called to replan the charging path, and the smart cabinet with the smallest deviation is prioritized according to the path deviation; S6: Output the final path to the mobile control system and simultaneously update the occupancy status of the smart cabinet database.

[0006] Optionally, the S1 specifically includes: S11: The mobile body receives the task instruction issued by the remote scheduling platform, including the target point coordinate information and task priority parameters, and then the mobile body parses the absolute position coordinates of the target point in the preset coordinate system. , and extract the task priority value ; S12: The controller inside the mobile object obtains the current lithium battery power information and calculates the current power state based on the voltage-capacity mapping relationship. The formula is: ,in, is the current power status, Indicates the current remaining power. Indicates the rated capacity of the lithium battery; S13: The mobile body conducts handshake communication with the surrounding smart cabinets through the nearest low-power wireless communication module, selects the smart cabinet with the largest signal strength as the target cabinet based on the preset RSSI threshold, and sends the target point coordinates to the smart cabinet. , Task priority value And the current power status Package it into an upload data structure and upload it to the status receiving end of the selected smart cabinet.

[0007] Optionally, the S2 specifically includes: S21: The smart cabinet activates a multimodal sensor array, including a lidar, a binocular camera, and an inertial measurement unit. The lidar acquires point cloud data of the surrounding environment; the binocular camera acquires synchronized stereo image pairs and reconstructs a 3D point cloud; and the inertial measurement unit collects current posture parameters. S22: Coordinate registration is performed on the point cloud data acquired by the lidar and the 3D point cloud of the binocular camera, and spatial alignment is performed based on the attitude parameters provided by the inertial measurement unit to form a preliminary sparse 3D point set; S23: Map the sparse 3D point set to a grid structure in 3D space, use a fixed voxel size for spatial division, record the point cloud density value within each voxel grid, and calculate the confidence score of the corresponding grid ; S24: Combine the grid structure and the corresponding confidence score and output it as a three-dimensional grid map.

[0008] Optionally, the confidence score The expression is: ,in, Indicates the The number of points in the grid, is the maximum number of points in all grids, Indicates the sensor matching quality score of the corresponding area of ​​the fused grid, 、 is the weighting coefficient, satisfying .

[0009] Optionally, the S3 specifically includes: S31: The mobile object reads the current power status obtained in S1 , and calculate the current available energy based on the rated capacity of the lithium battery ; S32: Obtain energy consumption per unit distance based on the lithium battery discharge model, and combine Calculate the maximum reachable radius of the moving object using the following formula: ,in, Indicates the maximum reachable radius; Indicates the power consumption per unit distance; S33: In the three-dimensional grid map generated by S2, the path length to each potential node is calculated using the graph search algorithm, starting from the current position of the mobile body. , and filter out The paths are taken as the initial candidate path set.

[0010] Optionally, the S4 specifically includes: S41: During the movement, the mobile object collects environmental data in real time through the on-board multimodal sensor array and compares the difference with the initial 3D grid map generated by S2 to form an environmental change data set ; S42: The mobile controller records the remaining power changes within a fixed time interval and calculates the power decay rate ; S43: Update the dataset based on the environment and battery decay rate , for each candidate path node To adjust the weight, the formula is: ,in, Indicates the The adjusted weights of the path nodes; Indicates based on Node environmental risk factors; Indicates based on The power attenuation factor; and is the weighting coefficient; S44: The moving body is adjusted according to the node weight Adjust path priorities and reorder the initial candidate path set.

[0011] Optionally, the S5 specifically includes: S51: The mobile body continuously evaluates the current remaining power and the energy consumption required to reach the next node on the path during the movement. If the current power is insufficient to support reaching the next node, path replanning is triggered; S52: Calling the smart cabinet location database, extracting the location information and occupancy status of all available smart cabinets in the current area, and mapping them to a three-dimensional grid map to form a charging target point set; S53: Based on the current path direction and the spatial position of each smart cabinet, an alternative path from the current position to each target smart cabinet is calculated, and the alternative path is compared with the original task path to calculate the path deviation, including the spatial deviation distance and the angular deviation; S54: Based on the charging target point set, compare the path deviations of all alternative charging options, sort them in ascending order of deviation, and preferentially select the smart cabinet with the smallest deviation and currently unoccupied as the target charging point.

[0012] Optionally, the S53 specifically includes: S531: After triggering path replanning, extract the location coordinates of all available target smart cabinets in the smart cabinet location database, and construct alternative paths from the current location to all available smart cabinet locations; S532: For each alternative path, extract the direction vector of the first path segment , calculate the direction vector of the original path The angle deviation between ; S533: Calculate the spatial deviation distance of each alternative path from the starting point to the target smart cabinet, and calculate the shortest Euclidean distance offset from the center line of the path from the original path to the final target, and use the weighted combination of the two as the spatial deviation index , the formula is: ,in, For the The spatial deviation distance indicator of the alternative path, Indicates the The path space deviation distance of the alternative path to the smart cabinet, Indicates the The shortest Euclidean distance offset between the alternative path and the original task path, is the weighting factor, satisfying ; The calculation formula is: ,in, For the The alternative path The spatial coordinates of the path nodes; For the The total number of nodes in the alternative paths; Represents the Euclidean distance between two points; The calculation formula is: ,in, From the current position to the A set of path points consisting of alternative paths of smart cabinets; is the set of path points of the original task path; For alternative paths a point on The original path a point on for point with dot The Euclidean distance between Indicates taking the minimum distance between all pairs of points; S534: Combined angle deviation Spatial deviation index , comprehensively calculate the path deviation as the ranking basis for smart cabinet path selection, the calculation formula is: ,in, represents the path deviation, is the weight coefficient.

[0013] Optionally, the S6 specifically includes: S61: After completing the path sequence sorting, the path with the lowest path deviation is selected as the final path, and a path instruction set including spatial coordinates, navigation instructions, and control parameters is generated according to the node sequence and encapsulated into a path control frame; S62: Sending the path control frame to the mobile control system via the internal communication bus; S63: After the path control frame is sent, the status of the smart cabinets involved in the selected path is registered, and the current occupancy status of the smart cabinet is recorded as occupied; S64: After receiving the occupancy status update request, the smart cabinet database updates the status flag of the target smart cabinet according to the uploaded timestamp and device number.

[0014] The mobile object path planning system based on multimodal sensing of smart cabinets and lithium battery power supply is used to implement the above-mentioned mobile object path planning method based on multimodal sensing of smart cabinets and lithium battery power supply, and includes the following modules: Task parsing module: used to receive task instructions sent by the remote scheduling platform, parse the target point coordinates and task priority parameters, and collect the current remaining power status information of the mobile body and transmit it to the path planning module; Environmental perception module: Deployed inside the smart cabinet, it includes a multimodal sensor array consisting of a lidar, a binocular camera, and an inertial measurement unit. It is used to collect surrounding environment data and generate a 3D grid map. Each voxel in the map is assigned a confidence score. Path Planning Module: This module connects the Task Analysis Module and the Environment Perception Module. It calculates the maximum reachable radius of the mobile object based on the current remaining battery status information, the three-dimensional grid map, and the lithium battery discharge model, and generates an initial set of candidate paths based on the target point coordinates. Dynamic adjustment module: Linked with the path planning module, it is used to receive the current position of the mobile body, environmental change data and power attenuation rate in real time, dynamically update the path node weights, and output the updated path sorting results; Charging path reconstruction module: When it detects that the remaining power is insufficient to support the next node, it calls the smart cabinet location database, calculates the path deviation between the current path and each candidate smart cabinet path, selects the smart cabinet with the smallest deviation to generate a charging path, and inserts it into the existing path; Path delivery and status synchronization module: connected to the dynamic adjustment module and the charging path reconstruction module, used to deliver the final path to the mobile control system and update the occupancy status of the selected smart cabinet to the smart cabinet database.

[0015] Beneficial effects of the present invention: The present invention achieves accurate identification and real-time updating of complex spatial obstacles through the fusion of environmental data by a multimodal sensor array, combined with a three-dimensional grid map and confidence score, effectively improving the accuracy and robustness of path planning for mobile objects in dynamic environments. Relying on the lithium battery discharge model and the remaining power status, the system can dynamically evaluate the maximum achievable radius and energy consumption boundaries, ensuring that path planning is always controlled by energy constraints.

[0016] The present invention, by introducing a smart cabinet location database and a path deviation evaluation mechanism, can automatically reconstruct the charging path when the mobile body is low on power, give priority to the smart cabinet with the smallest deviation, and achieve optimal insertion of the energy replenishment path; the system linkage path is synchronized with the smart cabinet status to ensure efficient coordination of task continuity and resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A schematic diagram of a mobile object path planning method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the process of selecting a smart cabinet with the smallest deviation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0020] It should be noted that references in the specification to "one embodiment," "an embodiment," "exemplary embodiments," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not necessarily every embodiment will include such specific features, structures, or characteristics. Furthermore, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0021] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0022] like Figure 1 As shown, the mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply includes the following steps: S1: The mobile object obtains the target point coordinates and task priority parameters, and simultaneously uploads the current state of charge (SOC) of the lithium battery to the nearest smart cabinet; S2: The smart cabinet collects environmental data through a multimodal sensor array (including lidar, binocular camera, and IMU) and generates a three-dimensional grid map with confidence scores; S3: Based on the current power state of S1, the three-dimensional grid map of S2, and the lithium battery discharge model, the maximum reachable radius of the mobile body is calculated to generate an initial set of candidate paths; S4: Real-time monitoring of environmental change data and battery attenuation rate of the moving object, and dynamic adjustment of path node weights; S5: When the remaining power cannot reach the next node, the smart cabinet location database is called to replan the charging path, and the smart cabinet with the smallest deviation is prioritized according to the path deviation; S6: Output the final path to the mobile control system and simultaneously update the occupancy status of the smart cabinet database.

[0023] S1 specifically includes: S11: The mobile body receives the task instruction issued by the remote scheduling platform, including the target point coordinate information and task priority parameters, and then the mobile body parses the absolute position coordinates of the target point in the preset coordinate system. , and extract the task priority value ; S12: The controller inside the mobile body obtains the current lithium battery power information and calculates the current state of charge (SOC) based on the voltage-capacity mapping relationship. The formula is: ,in, is the current power status, Indicates the current remaining power. Indicates the rated capacity of the lithium battery; S13: The mobile body conducts handshake communication with the surrounding smart cabinets through the nearest low-power wireless communication module (BLE or LoRa), selects the smart cabinet with the largest signal strength as the target cabinet based on the preset RSSI threshold, and writes the target point coordinates to the target cabinet. , Task priority value And the current power status The data is packaged into an upload data structure and uploaded to the status receiving end of the selected smart cabinet. The above steps ensure that the task information and energy status are accurately transmitted to the path planning node by explicitly adopting the task instruction parsing, voltage mapping method to calculate SOC, and the mechanism of selecting the optimal smart cabinet for uploading based on RSSI strength, providing basic support for subsequent path feasibility calculation and energy consumption evaluation.

[0024] S2 specifically includes: S21: The smart cabinet activates a multimodal sensor array, including a lidar, a binocular camera, and an inertial measurement unit. The lidar acquires point cloud data of the surrounding environment; the binocular camera acquires synchronized stereo image pairs and reconstructs a 3D point cloud; and the inertial measurement unit collects current posture parameters and acceleration information. The steps for obtaining a synchronized stereo image pair from a binocular camera and reconstructing a 3D point cloud are as follows: Synchronous image acquisition: The binocular camera module in the smart cabinet consists of two cameras, one for the left eye and one for the right eye. These two cameras capture images of the same scene at the same time through a hardware synchronization trigger mechanism, forming a synchronized image pair consisting of the left and right images. The optical axes of the two cameras are essentially parallel, with a fixed and known baseline distance (i.e., the relative position of the left and right cameras). Image correction and epipolar alignment: To eliminate lens distortion and imaging errors, the original image is dedistorted and epipolar correction is used to align the corresponding viewpoints in the left and right images to the same horizontal line, simplifying subsequent parallax calculations and improving matching accuracy. Disparity map construction: In the epipolar-aligned image pair, a disparity calculation method based on block matching or semi-global matching is used to search for correspondences between the left and right images pixel by pixel to obtain the disparity value of each pixel. The disparity reflects the horizontal displacement of the position of the point in the left and right images. A larger disparity indicates a closer target, while a smaller disparity indicates a farther target. 3D point cloud reconstruction: Using known camera intrinsic parameters (focal length, principal point position) and extrinsic parameters (baseline distance, posture), each image pixel with a valid disparity value is inversely calculated as a 3D point in space, obtaining the corresponding 3D coordinate point set. The spatial point positions of all pixels are combined to form 3D point cloud data; Coordinate transformation and output: The obtained three-dimensional point cloud is finally mapped to the unified world coordinate system of the smart cabinet, and the output is the three-dimensional perception result of the binocular vision module for fusion processing with other sensor data.

[0025] S22: Coordinate registration is performed on the point cloud data acquired by the lidar and the 3D point cloud of the binocular camera. Spatial alignment is performed based on the attitude parameters provided by the inertial measurement unit. The merged point cloud is downsampled using voxel filtering to form a preliminary sparse 3D point set. S23: Map the sparse 3D point set to a grid structure in 3D space, use a fixed voxel size for spatial division, record the point cloud density value within each voxel grid, and calculate the confidence score of the corresponding grid ; S24: Combining the grid structure with the corresponding confidence score, outputting the three-dimensional grid map, and storing the three-dimensional grid map in the local perception data cache of the smart cabinet for subsequent path planning.

[0026] Confidence score The expression is: ,in, Indicates the The number of points in the grid, is the maximum number of points in all grids, Indicates the sensor matching quality score of the corresponding area of ​​the fused grid, 、 is the weighting coefficient, satisfying .

[0027] S3 specifically includes: S31: The mobile object reads the current power status obtained in S1 , and calculate the current available energy based on the rated capacity of the lithium battery , the calculation formula is: ,in, Indicates the current available energy in Wh; Indicates the current power status in percentage; Indicates the rated capacity of the lithium battery in Wh; S32: Obtain energy consumption per unit distance based on the lithium battery discharge model, and combine Calculate the maximum reachable radius of the moving object using the following formula: ,in, Indicates the maximum reachable radius in units of ; Indicates the power consumption per unit distance, in units of ; The steps to obtain the energy consumption per unit distance based on the lithium battery discharge model are as follows: During the system initialization phase, the voltage-capacity characteristic curve of the lithium battery and the power consumption model of the mobile unit are preset. The overall average power is recorded as , unit is W; In actual operation, the moving body records the unit time through the wheel speed encoder Moving distance within ; Considering that the lithium battery voltage has nonlinear characteristics with SOC changes, according to the preset voltage-capacity curve, the instantaneous power value is corrected at the current SOC value to obtain the current actual energy consumption The correction method uses interpolation to match the power efficiency factor at the current voltage, resulting in: ,in, is the corresponding power efficiency factor; Divide the above actual energy consumption by the corresponding driving distance to obtain the energy consumption coefficient per unit distance: , whose expression is: ,in, The unit is , represents the amount of electricity required to travel 1 meter under the current load, voltage and motion status.

[0028] S33: In the three-dimensional grid map generated by S2, the path length to each potential node is calculated using the graph search algorithm, starting from the current position of the mobile body. , and filter out The path is taken as the initial candidate path set; the calculation formula is: ; ; ,in, Represents candidate paths The total length, in m; Indicates the path The distance of the segment, in units of ; Indicates the number of segments in the path; Represents the set of all candidate paths; is the maximum reachable radius currently calculated; by introducing the conversion mechanism between SOC and lithium battery capacity and combining it with the unit energy consumption model, the reachable radius range can be accurately evaluated and the path screening results can be ensured to have strict energy consumption constraints, providing basic support for the high-reliability operation of the path planning system.

[0029] S4 specifically includes: S41: During the movement, the mobile object collects environmental data in real time through the on-board multimodal sensor array and compares the difference with the initial 3D grid map generated by S2 to form an environmental change data set ; S42: The mobile controller records the remaining power changes within a fixed time interval and calculates the power decay rate , the calculation formula is: ,in, Indicates the rate of power decay, in Wh / s; Indicates the power reduction value during the time interval, in Wh; Indicates the recording time interval, the unit is s; S43: Update the dataset based on the environment and battery decay rate , for each candidate path node To adjust the weight, the formula is: ,in, Indicates the The adjusted weights of the path nodes; Indicates based on Node environmental risk factors; Indicates based on The power attenuation factor; and is the weighting coefficient, satisfying ; S44: The moving body is adjusted according to the node weight Adjust path priorities, reorder the initial candidate path set, and output a new path sequence for subsequent execution. Through coupled monitoring of real-time environmental changes and power attenuation rates and dynamic weight adjustment, these steps can timely correct path node priorities, effectively respond to environmental changes and energy consumption fluctuations, and improve the security and stability of path execution.

[0030] S5 specifically includes: S51: The mobile body continuously evaluates the current remaining power and the energy consumption required to reach the next node on the path during the movement. If the current power is insufficient to support reaching the next node, path replanning is triggered; S52: Calling the smart cabinet location database, extracting the location information and occupancy status of all available smart cabinets in the current area, and mapping them to a three-dimensional grid map to form a charging target point set; S53: Based on the current path direction and the spatial position of each smart cabinet, an alternative path from the current position to each target smart cabinet is calculated, and the alternative path is compared with the original task path to calculate the path deviation, including the spatial deviation distance and the angular deviation; S54: Based on the set of charging target points, compare the path deviations of all alternative charging options, sort them from small to large, and prioritize the smart cabinet with the smallest deviation and currently unoccupied as the target charging point; by building a path replanning mechanism triggered by power constraints, and combining the smart cabinet location database with a comprehensive evaluation of path deviations, it is possible to achieve minimal offset charging path insertion without interrupting the mission, ensuring the stable endurance and mission continuity of the mobile object in complex scenarios.

[0031] S53 specifically includes: S531: After triggering path replanning, obtain the current position of the moving object , the next reference node of the current path , calculate the original path direction vector At the same time, the location coordinates of all available target smart cabinets in the smart cabinet location database are extracted, and alternative paths from the current location to all available smart cabinet locations are constructed; S532: For each alternative path, extract the direction vector of the first path segment , calculate the direction vector of the original path The angle deviation between , the calculation formula is: ,in, Indicates the The direction deviation angle of the target smart cabinet is in radians; is the original path direction vector, is the direction vector of the initial segment of the replacement path, represents the vector dot product, Represents the vector modulus; S533: Calculate the spatial deviation distance of each alternative path from the starting point to the target smart cabinet, and calculate the shortest Euclidean distance offset from the center line of the path from the original path to the final target, and use the weighted combination of the two as the spatial deviation index , the formula is: ,in, For the The spatial deviation distance indicator of the alternative path, Indicates the path length from the current location to the smart cabinet, Indicates the shortest offset distance between this path and the original task path. is the weighting factor, satisfying ; S534: Combined angle deviation Spatial deviation index , comprehensively calculate the path deviation as the ranking basis for smart cabinet path selection, the calculation formula is: ,in, represents the path deviation, is the weight coefficient, satisfying By calculating the direction angle and spatial offset distance for each alternative path and constructing a unified path deviation quantification model, it is possible to accurately sort the candidate smart cabinet path solutions, ensuring that the selected path takes into account both charging feasibility and path continuity while minimizing the deviation from the original task path.

[0032] S6 specifically includes: S61: After completing the path sequence sorting, the path with the lowest path deviation is selected as the final path, and a path instruction set including spatial coordinates, navigation instructions, and control parameters is generated according to the node sequence and encapsulated into a path control frame; S62: Sending the path control frame to the mobile control system via the internal communication bus; S63: After the path control frame is sent, the status of the smart cabinets involved in the selected path is registered, the current occupancy status of the smart cabinet is recorded as occupied, and the status data is uploaded to the smart cabinet database; S64: After receiving the occupancy status update request, the smart cabinet database updates the status flag of the target smart cabinet according to the uploaded timestamp and device number to ensure that other scheduling requests will not repeatedly assign the smart cabinet, thereby completing the synchronous maintenance of the database; the above steps effectively realize the synchronous control of path execution and resource status by sending the final path to the mobile control system in the form of a structured control frame and synchronously updating the occupancy status of the smart cabinet database, ensuring the consistency of task scheduling and the mutual exclusivity of resource use, and providing a system foundation for path collaboration in multi-task scenarios.

[0033] like Figure 2 As shown, the mobile object path planning system based on multimodal perception of smart cabinets and lithium battery power supply is used to implement the above-mentioned mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply, including the following modules: Task parsing module: used to receive task instructions sent by the remote scheduling platform, parse the target point coordinates and task priority parameters, and collect the current remaining power status information of the mobile body and transmit it to the path planning module; Environmental perception module: Deployed inside the smart cabinet, it includes a multimodal sensor array consisting of a lidar, a binocular camera, and an inertial measurement unit. It is used to collect surrounding environment data and generate a 3D grid map. Each voxel in the map is assigned a confidence score. Path Planning Module: This module connects the Task Analysis Module and the Environment Perception Module. It calculates the maximum reachable radius of the mobile object based on the current remaining battery status information, the three-dimensional grid map, and the lithium battery discharge model, and generates an initial set of candidate paths based on the target point coordinates. Dynamic adjustment module: Linked with the path planning module, it is used to receive the current position of the mobile body, environmental change data and power attenuation rate in real time, dynamically update the path node weights, and output the updated path sorting results; Charging path reconstruction module: When it detects that the remaining power is insufficient to support the next node, it calls the smart cabinet location database, calculates the path deviation between the current path and each candidate smart cabinet path, selects the smart cabinet with the smallest deviation to generate a charging path, and inserts it into the existing path; Path delivery and status synchronization module: connected to the dynamic adjustment module and the charging path reconstruction module, used to deliver the final path to the mobile control system and update the occupancy status of the selected smart cabinet to the smart cabinet database.

[0034] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0035] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply, characterized in that: The following steps are involved: S1: The mobile object obtains the target point coordinates and task priority parameters, and simultaneously uploads the current power status of the lithium battery to the nearest smart cabinet; S2: The smart cabinet collects environmental data through a multimodal sensor array and generates a three-dimensional grid map with confidence scores; S3: Based on the current power state of S1, the three-dimensional grid map of S2, and the lithium battery discharge model, the maximum reachable radius of the mobile body is calculated to generate an initial set of candidate paths; S4: Real-time monitoring of environmental change data and battery attenuation rate of the moving object, and dynamic adjustment of path node weights; S5: When the remaining power cannot reach the next node, the smart cabinet location database is called to replan the charging path, and the smart cabinet with the smallest deviation is prioritized according to the path deviation; S6: Output the final path to the mobile control system and simultaneously update the occupancy status of the smart cabinet database.

2. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 1 is characterized in that: Said S1 specifically includes: S11: The mobile body receives the task instruction issued by the remote scheduling platform, including the target point coordinate information and task priority parameters, and then the mobile body parses the absolute position coordinates of the target point in the preset coordinate system. , and extract the task priority value ; S12: The controller inside the mobile object obtains the current lithium battery power information and calculates the current power state based on the voltage-capacity mapping relationship. The formula is: ,in, is the current power status, Indicates the current remaining power. Indicates the rated capacity of the lithium battery; S13: The mobile body conducts handshake communication with the surrounding smart cabinets through the nearest low-power wireless communication module, selects the smart cabinet with the largest signal strength as the target cabinet based on the preset RSSI threshold, and sends the target point coordinates to the smart cabinet. , Task priority value And the current power status Package it into an upload data structure and upload it to the status receiving end of the selected smart cabinet.

3. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 1 is characterized in that: The S2 specifically includes: S21: The smart cabinet activates a multimodal sensor array, including a lidar, a binocular camera, and an inertial measurement unit. The lidar acquires point cloud data of the surrounding environment; the binocular camera acquires synchronized stereo image pairs and reconstructs a 3D point cloud; and the inertial measurement unit collects current posture parameters. S22: Coordinate registration is performed on the point cloud data acquired by the lidar and the 3D point cloud of the binocular camera, and spatial alignment is performed based on the attitude parameters provided by the inertial measurement unit to form a preliminary sparse 3D point set; S23: Map the sparse 3D point set to a grid structure in 3D space, use a fixed voxel size for spatial division, record the point cloud density value within each voxel grid, and calculate the confidence score of the corresponding grid ; S24: Combine the grid structure and the corresponding confidence score and output it as a three-dimensional grid map.

4. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 3 is characterized in that: The confidence score The expression is: ,in, Indicates the The number of points in the grid, is the maximum number of points in all grids, Indicates the sensor matching quality score of the corresponding area of ​​the fused grid, 、 is the weighting coefficient, satisfying .

5. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 1 is characterized in that: The S3 specifically includes: S31: The mobile object reads the current power status obtained in S1 , and calculate the current available energy based on the rated capacity of the lithium battery ; S32: Obtain energy consumption per unit distance based on the lithium battery discharge model, and combine Calculate the maximum reachable radius of the moving object using the following formula: ,in, Indicates the maximum reachable radius; Indicates the power consumption per unit distance; S33: In the three-dimensional grid map generated by S2, the path length to each potential node is calculated using the graph search algorithm, starting from the current position of the mobile body. , and filter out The paths are taken as the initial candidate path set.

6. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 1 is characterized in that: The S4 specifically includes: S41: During the movement, the mobile object collects environmental data in real time through the on-board multimodal sensor array and compares the difference with the initial 3D grid map generated by S2 to form an environmental change data set ; S42: The mobile controller records the remaining power changes within a fixed time interval and calculates the power decay rate ; S43: Update the dataset based on the environment and battery decay rate , for each candidate path node To adjust the weight, the formula is: ,in, Indicates the The adjusted weights of the path nodes; Indicates based on Node environmental risk factors; Indicates based on The power attenuation factor; and is the weighting coefficient; S44: The moving body is adjusted according to the node weight Adjust path priorities and reorder the initial candidate path set.

7. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 1 is characterized in that: The S5 specifically includes: S51: The mobile body continuously evaluates the current remaining power and the energy consumption required to reach the next node on the path during the movement. If the current power is insufficient to support reaching the next node, path replanning is triggered; S52: Calling the smart cabinet location database, extracting the location information and occupancy status of all available smart cabinets in the current area, and mapping them to a three-dimensional grid map to form a charging target point set; S53: Based on the current path direction and the spatial position of each smart cabinet, an alternative path from the current position to each target smart cabinet is calculated, and the alternative path is compared with the original task path to calculate the path deviation, including the spatial deviation distance and the angular deviation; S54: Based on the charging target point set, compare the path deviations of all alternative charging options, sort them in ascending order of deviation, and preferentially select the smart cabinet with the smallest deviation and currently unoccupied as the target charging point.

8. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 7 is characterized in that: The S53 specifically includes: S531: After triggering path replanning, extract the location coordinates of all available target smart cabinets in the smart cabinet location database, and construct alternative paths from the current location to all available smart cabinet locations; S532: For each alternative path, extract the direction vector of the first path segment , calculate the direction vector of the original path The angle deviation between ; S533: Calculate the spatial deviation distance of each alternative path from the starting point to the target smart cabinet, and calculate the shortest Euclidean distance offset from the center line of the path from the original path to the final target, and use the weighted combination of the two as the spatial deviation index ; S534: Combined angle deviation Spatial deviation index , comprehensively calculate the path deviation as the ranking basis for smart cabinet path selection, the calculation formula is: ,in, represents the path deviation, is the weight coefficient.

9. The mobile object path planning method based on multimodal perception of smart cabinets and lithium battery power supply according to claim 1 is characterized in that: The S6 specifically includes: S61: After completing the path sequence sorting, the path with the lowest path deviation is selected as the final path, and a path instruction set including spatial coordinates, navigation instructions, and control parameters is generated according to the node sequence and encapsulated into a path control frame; S62: Sending the path control frame to the mobile control system via the internal communication bus; S63: After the path control frame is sent, the status of the smart cabinets involved in the selected path is registered, and the current occupancy status of the smart cabinet is recorded as occupied; S64: After receiving the occupancy status update request, the smart cabinet database updates the status flag of the target smart cabinet according to the uploaded timestamp and device number.

10. A mobile object path planning system based on multimodal sensing of a smart cabinet and lithium battery power supply, for implementing a mobile object path planning method based on multimodal sensing of a smart cabinet and lithium battery power supply as described in any one of claims 1 to 9, characterized in that: Includes the following modules: Task parsing module: used to receive task instructions sent by the remote scheduling platform, parse the target point coordinates and task priority parameters, and collect the current remaining power status information of the mobile body and transmit it to the path planning module; Environmental perception module: Deployed inside the smart cabinet, it includes a multimodal sensor array consisting of a lidar, a binocular camera, and an inertial measurement unit. It is used to collect surrounding environment data and generate a 3D grid map. Each voxel in the map is assigned a confidence score. Path Planning Module: This module connects the Task Analysis Module and the Environment Perception Module. It calculates the maximum reachable radius of the mobile object based on the current remaining battery status information, the three-dimensional grid map, and the lithium battery discharge model, and generates an initial set of candidate paths based on the target point coordinates. Dynamic adjustment module: Linked with the path planning module, it is used to receive the current position of the mobile body, environmental change data and power attenuation rate in real time, dynamically update the path node weights, and output the updated path sorting results; Charging path reconstruction module: When it detects that the remaining power is insufficient to support the next node, it calls the smart cabinet location database, calculates the path deviation between the current path and each candidate smart cabinet path, selects the smart cabinet with the smallest deviation to generate a charging path, and inserts it into the existing path; Path delivery and status synchronization module: connected to the dynamic adjustment module and the charging path reconstruction module, used to deliver the final path to the mobile control system and update the occupancy status of the selected smart cabinet to the smart cabinet database.

Images