Automatic driving target positioning method and system based on multi-source data fusion

Abstract

The invention discloses an automatic driving target positioning method and system based on multi-source data fusion, and relates to the technical field of data fusion. The method comprises the following steps: performing multi-source sensing fusion on a to-be-charged object of the mobile storage and charging equipment to obtain a target sensing sequence; performing multi-level positioning on the to-be-charged object to obtain a first target positioning feature and a second target positioning feature; performing perception interference correction to obtain a third target positioning feature and a fourth target positioning feature; performing driving level analysis on the mobile storage and charging equipment, and constructing a driving level factor; driving control decision making is carried out on the mobile storage and charging device to obtain a first driving control space, iterative optimization is carried out on the first driving control space, and a driving control strategy is determined; and carrying out driving management on the mobile storage and charging equipment, and carrying out updating driving control in combination with a multi-source sensor. The technical problem of inaccurate target positioning of automatic driving mobile storage and charging equipment in the prior art is solved, and the technical effect of improving the positioning accuracy is achieved.

Description

Technical Field

[0001] The present invention relates to the field of data fusion technology, and in particular to an autonomous driving target positioning method and system based on multi-source data fusion. Background Art

[0002] In smart electric vehicle charging scenarios, mobile charging and storage equipment can be flexibly moved to charging locations, providing convenient charging services for new energy vehicles and effectively alleviating issues such as insufficient fixed charging station layouts and long charging queues. However, existing mobile charging and storage equipment typically relies on a single type of sensor for environmental perception and target positioning during autonomous driving. Due to limited sensor sources and susceptibility to factors such as ambient lighting, occlusion, and noise interference, this results in insufficient target positioning accuracy, charging object identification bias, unstable path planning, and poor reliability of driving control strategies. This not only reduces the operational efficiency of autonomous driving charging and storage equipment in complex scenarios, but also affects the overall safety and stability of operation. Summary of the Invention

[0003] This application provides an autonomous driving target positioning method and system based on multi-source data fusion, which solves the technical problem in the prior art that autonomous driving mobile storage and charging equipment has inaccurate target positioning due to a single perception method.

[0004] In a first aspect of the present application, a method for autonomous driving target positioning based on multi-source data fusion is provided, the method comprising: A multi-source perception fusion is performed on the object to be charged of the mobile charging and storage device through a multi-source sensor to obtain a target perception sequence; a multi-level positioning is performed on the object to be charged according to the target perception sequence to obtain a first target positioning feature and a second target positioning feature; a perception interference correction is performed on the first target positioning feature and the second target positioning feature based on the multi-source sensor to obtain a third target positioning feature and a fourth target positioning feature; a driving hierarchy analysis is performed on the mobile charging and storage device according to the third target positioning feature and the fourth target positioning feature to construct a driving hierarchy factor; a driving control decision is made on the mobile charging and storage device based on the driving hierarchy factor to obtain a first driving control space, and the first driving control space is iteratively optimized according to a driving risk analysis model to determine a driving control strategy; driving management of the mobile charging and storage device is performed based on the driving control strategy, and driving control is updated in combination with the multi-source sensor.

[0005] A second aspect of the present application provides an autonomous driving target positioning system based on multi-source data fusion, the system comprising: Multi-source perception fusion module: performs multi-source perception fusion on the object to be charged of the mobile storage and charging device through a multi-source sensor to obtain a target perception sequence; multi-level positioning module: performs multi-level positioning on the object to be charged according to the target perception sequence to obtain a first target positioning feature and a second target positioning feature; perception interference correction module: performs perception interference correction on the first target positioning feature and the second target positioning feature based on the multi-source sensor to obtain a third target positioning feature and a fourth target positioning feature; driving level analysis module: performs driving level analysis on the mobile storage and charging device according to the third target positioning feature and the fourth target positioning feature to construct a driving level factor; driving control decision module: performs driving control decision on the mobile storage and charging device based on the driving level factor to obtain a first driving control space, and iteratively optimizes the first driving control space according to the driving risk analysis model to determine a driving control strategy; driving management module: performs driving management on the mobile storage and charging device based on the driving control strategy, and updates the driving control in combination with the multi-source sensor.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, a multi-source sensor is used to perform multi-source perception fusion on the charging object of a mobile charging and storage device to obtain a target perception sequence. Next, a multi-level positioning of the charging object is performed based on the target perception sequence, obtaining a first target positioning feature and a second target positioning feature. Furthermore, perceptual interference correction is performed on the first and second target positioning features based on the multi-source sensor to obtain a third and fourth target positioning feature. Next, a driving hierarchy analysis is performed on the mobile charging and storage device based on the third and fourth target positioning features to construct a driving hierarchy factor. Then, driving control decisions are made for the mobile charging and storage device based on the driving hierarchy factor to obtain a first driving control space. This first driving control space is then iteratively optimized using a driving risk analysis model to determine a driving control strategy. Finally, driving management of the mobile charging and storage device is performed based on the driving control strategy, and driving control is updated in conjunction with the multi-source sensor. This method solves the technical problem of inaccurate target positioning in autonomous mobile charging and storage devices due to a single perception method in the prior art, achieving the technical effect of improving target positioning accuracy through multi-source data fusion and interference correction. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A flowchart of a method for autonomous driving target positioning based on multi-source data fusion provided in an embodiment of the present application; Figure 2 Schematic diagram of the structure of the autonomous driving target positioning system based on multi-source data fusion provided in an embodiment of the present application.

[0009] Explanation of the reference numerals: multi-source perception fusion module 11 , multi-level positioning module 12 , perception interference correction module 13 , driving level analysis module 14 , driving control decision module 15 , driving management module 16 . DETAILED DESCRIPTION

[0010] This application solves the technical problem in the prior art of inaccurate target positioning of autonomous driving mobile storage and charging equipment due to a single perception method by providing an autonomous driving target positioning method and system based on multi-source data fusion.

[0011] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0012] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0013] Example 1, as Figure 1 As shown, the present application provides an autonomous driving target positioning method based on multi-source data fusion, wherein the method includes: The multi-source sensor is used to perform multi-source perception fusion on the charging object of the mobile charging and storage device to obtain the target perception sequence.

[0014] The multi-source sensor can be composed of a laser radar, a millimeter wave radar, an infrared imaging sensor, a depth camera, and an ultrasonic sensor, and is used for omnidirectional and heterogeneous environmental perception of the object to be charged during driving of the mobile charging device. The laser radar is used to collect three-dimensional point cloud data of the object to be charged, and obtain the spatial geometric shape and relative distance of the object. The millimeter wave radar is used to obtain the speed information and motion trajectory of the object in a low-visibility or insufficient-light scene. The infrared imaging sensor is used to capture the thermal features of the object in the night or complex light conditions, and realize thermal signal compensation. The depth camera is used to collect two-dimensional images and depth information of the object, and enhance the geometric contour recognition capability. The ultrasonic sensor is used for boundary detection and obstacle recognition in a short distance range.

[0015] By synchronizing the time stamps of the original data streams collected by the multi-source sensor and cleaning the data, redundant and noise information is removed to obtain a multi-source perception effective data set. The multi-source perception effective data set is subjected to feature extraction and space-time alignment, and the three-dimensional point cloud features, motion trajectory features, thermal features, image features, and boundary features are weighted and fused according to a preset fusion algorithm to generate a target perception sequence. The target perception sequence is used to represent the multi-dimensional perception information of the spatial position, attitude, motion trend, and charging interface region of the object to be charged.

[0016] Further, the multi-source perception fusion of the object to be charged of the mobile charging device by the multi-source sensor obtains the target perception sequence, including: According to the real-time perception of the object to be charged by the multi-source sensor, a target perception first set is obtained. The target perception first set is cleaned to obtain a target perception second set. The target perception second set is fused to generate the target perception sequence.

[0017] Based on the real-time perception of the object to be charged by the multi-source sensor, the spatial three-dimensional point cloud information, motion speed information, appearance image information, and short-distance boundary information of the object to be charged are collected to form a target perception first set. The target perception first set is subjected to data cleaning processing, specifically including: time stamp alignment and spatial coordinate system unification of different perception source data, removal of abnormal values, noise points, and redundant information, and correction of missing data by interpolation compensation and filtering algorithm to generate a target perception second set with higher quality. The target perception second set is subjected to multi-source fusion processing. The fusion method can be feature-level fusion based on weight distribution. The fusion result can comprehensively reflect the spatial position, attitude feature, charging interface position, and dynamic change trend of the object to be charged, thereby generating the target perception sequence.

[0018] Feature-level fusion based on weight allocation includes: extracting features from sensory data from lidar, millimeter-wave radar, camera, infrared sensor and ultrasonic sensor to obtain point cloud features, velocity features, image features, thermal imaging features and boundary features respectively; setting corresponding weight factors according to the reliability of different sensors in positioning accuracy, environmental adaptability and anti-interference ability, for example, assigning a higher spatial geometry weight to lidar point cloud features, assigning dynamic tracking weights to millimeter-wave radar motion features, and assigning appearance recognition and illumination compensation weights to cameras and infrared sensors; then, under a unified time coordinate and space coordinate system, the above features are weightedly superimposed or fused according to their respective weights to obtain a fused multi-dimensional comprehensive feature vector.

[0019] The object to be charged is positioned at multiple levels according to the target sensing sequence to obtain a first target positioning feature and a second target positioning feature.

[0020] Furthermore, the first target positioning feature includes the geographical location of the object to be charged, and the second target positioning feature includes the location of the charging port of the object to be charged.

[0021] Based on the three-dimensional point cloud features and motion trajectory features obtained by the lidar and millimeter-wave radar in the target perception sequence, combined with the global positioning system (such as GNSS / IMU) data of the mobile storage and charging equipment, the global spatial position of the charging object is estimated, thereby determining its absolute coordinates within the road or station, and obtaining the first target positioning feature, that is, the geographic location of the charging object.

[0022] Based on the images and thermal features collected by the depth camera and infrared sensor in the target perception sequence, the geometric shape of the object to be charged is segmented and local area recognition is performed, the charging interface area of ​​the object to be charged is identified, and the charging interface area is finely positioned in a unified coordinate system. The spatial position of the charging interface is obtained through edge detection and area matching, and the second feature of target positioning, namely the charging interface position of the object to be charged, is obtained.

[0023] Based on the multi-source sensor, perceptual interference correction is performed on the first target positioning feature and the second target positioning feature to obtain a third target positioning feature and a fourth target positioning feature.

[0024] Furthermore, performing perceptual interference correction on the first target positioning feature and the second target positioning feature based on the multi-source sensor to obtain a third target positioning feature and a fourth target positioning feature includes: The multi-source perception device is subjected to perception scene collection based on the target perception sequence, to obtain a plurality of perception scene data; interference factors of the multi-source perception device are identified based on the plurality of perception scene data, to obtain a perception interference identification result; the target positioning first feature is subjected to positioning error correction based on the perception interference identification result, to obtain a target positioning third feature; and the target positioning second feature is subjected to positioning error correction based on the perception interference identification result, to obtain a target positioning fourth feature.

[0025] The multi-source perception device is subjected to perception scene collection based on the target perception sequence, specifically, during the operation of the mobile storage and charging device, sensors such as laser radars, millimeter wave radars, cameras, depth sensors, and infrared sensors are used to continuously collect environmental data of the object to be charged under different time periods, different light conditions, and different spatial positions, to form a plurality of perception scene data sets; the output results of the multi-source perception device are subjected to interference factor identification based on the plurality of perception scene data, wherein the interference factors include data abnormalities caused by insufficient light, existence of an obstacle, reflection interference, electromagnetic noise, and multipath effect, and the interference factor identification can be achieved through feature residual detection, statistical threshold determination, or an interference classification model based on deep learning, to obtain a perception interference identification result; the target positioning first feature is subjected to positioning error correction based on the perception interference identification result, for example, the positioning result is corrected through Kalman filtering, weighted average, or least square estimation, to obtain a target positioning third feature; and the target positioning second feature is subjected to positioning error correction based on the perception interference identification result, for example, the positioning result is corrected through edge detection correction, region matching optimization, or an interface recognition enhancement model based on a convolutional neural network, to obtain a target positioning fourth feature.

[0026] The mobile storage and charging device is subjected to driving level analysis based on the target positioning third feature and the target positioning fourth feature, to construct a driving level factor.

[0027] The third target positioning feature is used to determine the geographical position of the object to be charged in the global environment, and a macroscopic navigation path from the current position of the device to the target position is constructed in combination with the current geographical coordinates of the mobile storage and charging device. The fourth target positioning feature is used to determine the accurate position of the charging interface of the object to be charged in space, and a microscopic docking path is established in combination with the three-dimensional structural features of the target object. The macroscopic navigation path is used as the first driving level, and the microscopic docking path is used as the second driving level. The two are hierarchically analyzed to obtain a multi-level driving task from global driving to accurate docking. On this basis, the path elements from the current position of the device to the target position are used as the first driving factor, and the path elements from the external docking position of the target object to the position of the charging interface are used as the second driving factor. The two driving factors are modified and normalized according to the constraint conditions such as the distribution of environmental obstacles, the turning radius of the device, and the docking tolerance range, and finally the driving level factor is constructed.

[0028] Further, the driving level analysis of the mobile storage and charging device is performed according to the third target positioning feature and the fourth target positioning feature, and the driving level factor is constructed, including: According to the three-dimensional structural features of the object to be charged and the space features, a target three-dimensional space is constructed. Based on the target three-dimensional space, the end point tolerance fitting of the mobile storage and charging device is performed according to the third target positioning feature, and the end point tolerance region is obtained. The end point tolerance region is corrected according to the device structural features of the mobile storage and charging device, and the end point tolerance position is generated. The device geographical position of the mobile storage and charging device to the end point tolerance position is used as the first driving factor. The external charging position of the device corresponding to the end point tolerance position to the fourth target positioning feature is used as the second driving factor, and the first driving factor and the second driving factor are added to the driving level factor.

[0029] Based on the three-dimensional structural features of the object to be charged (such as the geometric shape and size parameters of the vehicle or device) and the space features (such as the parking area boundary, the distribution of surrounding obstacles, and the ground markings), a target three-dimensional space is constructed to describe the spatial topological relationship between the object to be charged and the surrounding environment.

[0030] In the target three-dimensional space, based on the geographic location of the charging object represented by the third feature of target positioning, a tolerance fitting is performed on the navigation destination of the mobile charging and storage device. Specifically, a candidate region of the destination is established within the charging docking area. By setting a position deviation threshold and docking tolerance parameters, a tolerance domain of the destination is obtained. The tolerance domain of the destination is corrected for reachability based on the structural characteristics of the mobile charging and storage device (such as the minimum turning radius, overall dimensions, and the telescopic range of the docking arm). Candidate regions that fail to meet the structural constraints are eliminated, thereby generating a tolerance location for the destination. The path element from the geographic location of the mobile charging and storage device to the tolerance location of the destination is used as the first driving factor to describe the driving path at the macro level. The path element from the device external charging location (the interface location on the mobile charging and storage device for energy exchange with the charging object) corresponding to the tolerance location of the destination is used as the second driving factor to describe the precise docking path at the micro level. The first and second driving factors are integrated and added to the driving hierarchy factor, allowing the driving hierarchy factor to simultaneously reflect the hierarchical control requirements of global navigation and local docking.

[0031] A driving control decision is made for the mobile charging and storage device based on the driving hierarchy factor to obtain a first driving control space, and the first driving control space is iteratively optimized according to a driving risk analytical model to determine a driving control strategy.

[0032] Based on the multimodal control constraint information of the mobile charging and storage device, including speed constraints, steering constraints, path smoothness constraints, and device docking accuracy constraints, driving control constraints are constructed. Based on the driving control constraints, a multi-objective decision analysis is performed on the driving hierarchical factors to generate multiple feasible driving control candidate solutions. These solutions are then combined to form a first driving control space. A driving risk analysis model is used to assess the risk of each candidate solution in the first driving control space. The driving risk analysis model includes multi-dimensional risk assessment indicators such as driving safety risk, docking anomaly risk, and driving failure risk, and outputs a corresponding risk constraint matrix. Based on the risk constraint matrix, an iterative optimization search is performed on the first driving control space, gradually eliminating candidate solutions that do not meet safety and reliability requirements. The remaining solutions are then risk-ranked and optimized to obtain a driving control strategy that meets driving safety, docking accuracy, and operational efficiency.

[0033] Furthermore, making a driving control decision for the mobile charging and storage device based on the driving level factor to obtain a first driving control space includes: Driving scene features are collected based on the first driving factor and the second driving factor to construct a first driving scene model; obstacle features are labeled based on the first driving scene model to obtain a second driving scene model; multimodal driving control constraint information of the mobile charging and storage device is collected to generate driving control constraint conditions; based on the driving control constraint conditions, a driving control decision is made for the first driving factor according to the second driving scene model to obtain a first driving control decision set; based on the driving control constraint conditions, a driving control decision is made for the second driving factor according to the second driving scene model to obtain a second driving control decision set; and the first driving control decision set and the second driving control decision set are randomly combined to generate the first driving control space.

[0034] Based on the first driving factor and the second driving factor, features of the environmental perception data are collected, and the road boundaries, environmental obstacle distribution, and geometric constraints of the target area are extracted to construct a preliminary first driving scene model. In the first driving scene model, potential obstacles are feature-labeled. For example, through lidar point cloud clustering, camera target recognition, or millimeter-wave radar dynamic tracking, the position, size, and motion trend of the obstacles are labeled and classified to obtain a second driving scene model that has been enhanced with obstacles.

[0035] Collect multimodal driving control constraint information of mobile storage and charging equipment, including speed limit, turning radius, acceleration and deceleration smoothness, path curvature threshold, and docking terminal posture constraint, and generate driving control constraint conditions based on this.

[0036] Based on driving control constraints and in combination with the second driving scenario model, control decisions are made for the first driving factor to generate a first driving control decision set. The first driving control decision set is used to describe multiple feasible paths and control instructions for the device from its current location to the tolerance location of the final destination. Simultaneously, based on the same constraints and the second driving scenario model, control decisions are made for the second driving factor to generate a second driving control decision set. The second driving control decision set is used to describe multiple refined docking paths and control instructions from the external charging position to the charging port position during the device terminal docking process. The first and second driving control decision sets are randomly combined to generate a first driving control space that covers the set of feasible control solutions for driving factors at different levels under constraints.

[0037] Furthermore, iteratively optimizing the first driving control space according to the driving risk analytical model to determine the driving control strategy includes: Constraints are configured based on multidimensional driving risk evaluation indicators of the driving risk analysis model to generate a driving risk constraint matrix, where the multidimensional driving risk evaluation indicators include driving safety risk, docking abnormality risk, and driving failure risk; based on the driving risk analysis model, driving risk constraint optimization is performed on the first driving control space according to the driving risk constraint matrix to establish a second driving control space; weights are configured based on the multidimensional driving risk evaluation indicators to obtain a comprehensive driving risk analysis model; comprehensive driving risk ranking optimization is performed on the second driving control space according to the comprehensive driving risk analysis model to obtain K driving control candidate solutions, where K is a positive integer; based on the driving risk analysis model and the comprehensive driving risk analysis model, variation optimization is performed on the K driving control candidate solutions to establish a third driving control space; and driving time minimization optimization is performed on the third driving control space to generate the driving control strategy.

[0038] The driving risk analysis model is a multi-dimensional risk assessment model based on machine learning, which is used to quantitatively assess various types of risks that may occur during driving. Optionally, the operating data of the mobile storage and charging equipment in different driving scenarios is collected, including vehicle driving trajectory data, speed and acceleration data, obstacle approach distance data, charging interface docking deviation data, sensor status data, and control execution feedback data; a multi-dimensional risk label system is established based on the operating data, and the risk labels include at least a driving safety risk label, a docking abnormality risk label, and a driving failure risk label, wherein the driving safety risk label is used to mark events in which the vehicle collides or deviates during driving, the docking abnormality risk label is used to mark events in which offset or docking failure occurs during the charging interface identification and docking process, and the driving failure risk label is used to mark driving errors caused by sensor failure or controller abnormality. The operating data is subjected to feature engineering processing, including data cleaning, time series segmentation, feature extraction and normalization, to obtain a high-dimensional feature vector that characterizes the relationship between the vehicle's operating status and risk events. On this basis, a machine learning method is used to model the high-dimensional feature vector and risk labels. Random forest models, support vector machine models, deep neural network models or integrated learning models can be used for training to obtain a driving risk analysis model that can simultaneously predict driving safety risks, docking abnormality risks and driving failure risks.

[0039] The multidimensional driving risk assessment indicators in the driving risk analysis model are used, including driving safety risk, docking anomaly risk, and driving failure risk. Constraints are set for each risk assessment indicator, such as a collision probability threshold constraint for driving safety risk, a docking deviation range constraint for docking anomaly risk, and an operational reliability threshold constraint for driving failure risk. These constraints are aggregated to form a driving risk constraint matrix. This driving risk constraint matrix is ​​applied to the first driving control space, and multiple driving control solutions are screened for risk constraints one by one. Solutions that do not meet the constraints are eliminated, resulting in a second driving control space that meets the constraint requirements.

[0040] Based on the multi-dimensional driving risk evaluation indicators, weights are configured to construct a comprehensive driving risk analysis model. That is, each risk indicator is weighted according to the preset importance ratio to construct a comprehensive driving risk function, which is used to calculate the comprehensive risk of any driving control scheme and output the comprehensive driving risk coefficient.

[0041] Driving comprehensive risk function: , where R is the comprehensive driving risk coefficient, is the measurement value of the risk assessment index of type i, is the corresponding weight factor. The comprehensive driving risk function can be used to perform a unified quantitative evaluation of all candidate solutions in the second driving control space. The comprehensive driving risk coefficient is used as the sorting criterion, and the top K driving control candidate solutions are sorted from small to large, selecting them.

[0042] Based on the driving risk analysis model and the comprehensive driving risk analysis model, K candidate driving control schemes are identified and optimized through differences. A new driving variation group is generated through crossover and difference operators. After risk constraint correction and comprehensive risk re-ranking, a third driving control space is established. With the goal of minimizing driving time in the third driving control space, multiple schemes in the third driving control space are iteratively optimized and path fitting is performed to select a driving control scheme that meets the requirements of safety, reliability, and time optimization as the driving control strategy.

[0043] Furthermore, based on the driving risk analytical model, the first driving control space is optimized according to the driving risk constraint matrix to establish a second driving control space, including: A Wth driving control scheme is extracted based on the first driving control space, where W is a positive integer; a Wth simulated driving data is obtained by simulated driving of the mobile charging and storage device based on the Wth driving control scheme; the Wth simulated driving data is input into the driving risk analysis model to obtain a Wth driving risk matrix; if the Wth driving risk matrix satisfies the driving risk constraint matrix, the Wth driving control scheme is added to the second driving control space; and a driving risk constraint optimization is continued for the first driving control space based on the driving risk analysis model and the driving risk constraint matrix to obtain the second driving control space.

[0044] A Wth driving control scheme is extracted from the first driving control space; a mobile charging device is driven to perform simulated driving based on the Wth driving control scheme, and dynamic data of the vehicle during navigation path, speed change, obstacle avoidance, and docking is collected to generate Wth simulated driving data; the Wth simulated driving data is input into a driving risk analysis model, and risk is calculated using multi-dimensional driving risk evaluation indicators within the model to output a Wth driving risk matrix, wherein the driving risk matrix includes quantified results of driving safety risk, docking abnormality risk, and driving failure risk; the Wth driving risk matrix is ​​compared with a preset driving risk constraint matrix; if the Wth driving control scheme satisfies all constraints, the corresponding Wth driving control scheme is added to the second driving control space; if not, the corresponding Wth driving control scheme is removed; the above process is repeated, and the remaining schemes in the first driving control space are simulated, calculated, and screened in sequence until the risk constraint verification of all schemes is completed, thereby obtaining a second driving control space that satisfies the requirements of the driving risk constraint matrix.

[0045] Furthermore, based on the driving risk analysis model and the driving comprehensive risk analysis model, the K driving control candidate solutions are subjected to mutation optimization to establish a third driving control space, including: The second driving control space is subjected to difference identification based on the K driving control candidate solutions to obtain a plurality of driving control difference vectors; based on the driving hierarchy factor, the second driving control space is subjected to cross-mutation based on the plurality of driving control difference vectors to obtain a first driving variation group; based on the driving risk analytical model, the first driving variation group is subjected to driving risk constraint optimization according to the driving risk constraint matrix to establish a second driving variation group; the second driving variation group is expanded based on the K driving control candidate solutions to obtain a third driving variation group; and the third driving variation group is subjected to driving comprehensive risk ranking optimization according to the driving comprehensive risk analytical model to obtain the third driving control space.

[0046] The second driving control space is identified according to K driving control candidate schemes, specifically, the candidate schemes are compared in terms of path length, steering angle distribution, speed change curve and end-to-end docking attitude, and a plurality of driving control difference vectors are obtained to represent the feature differences between the candidate schemes.

[0047] Based on the driving level factor, the plurality of driving control difference vectors are introduced into the crossover mutation operator to perform crossover combination and disturbance on the second driving control space, to generate a new driving control scheme group, to obtain a driving variation first group, thereby expanding the diversity and coverage of the original candidate scheme. Based on the driving risk analysis model, the driving variation first group is risk screened and optimized according to a preset driving risk constraint matrix, to eliminate the control schemes that do not satisfy the safety threshold, to establish a driving variation second group; the driving variation second group is expanded according to the initial K driving control candidate schemes, so as to have the mixed characteristics of the original candidate schemes and the variation schemes, thereby obtaining a driving variation third group; based on the driving comprehensive risk analysis model, all schemes in the driving variation third group are driving comprehensive risk sorted and optimized, and are sequentially sorted in order of comprehensive risk value from small to large, and the top K driving control schemes are selected and added to the third driving control space, to obtain the third driving control space after variation and optimization.

[0048] Based on the driving control strategy, the mobile storage and charging equipment is driven and managed, and the multi-source sensor is combined to update the driving control.

[0049] The driving control strategy generated by iterative optimization is issued to the execution unit of the mobile storage and charging equipment, and the driving path planning, speed control, steering adjustment and charging interface docking operation of the vehicle are managed in real time, so as to ensure that the mobile storage and charging equipment automatically drives and docks according to the optimal control strategy.

[0050] During the driving process of the mobile storage and charging equipment, the multi-source sensor continuously dynamically perceives the to-be-charged object and the surrounding environment, collects updated perception data, including the real-time position offset of the to-be-charged object, the attitude change of the charging interface and the dynamic change of the environmental obstacles; then, the updated perception data and the current driving control strategy are compared, if there is a significant deviation or environmental sudden change, the driving control update process is triggered, that is, the updated perception data is taken as input, the target perception sequence generation, multi-level positioning, interference correction, driving level analysis, control space construction and risk optimization are repeated, a new driving control strategy is generated in real time, and the original control instruction is replaced by the updated driving control strategy and continues to be executed, so as to realize the dynamic closed loop of the driving management process.

[0051] In summary, the embodiments of the present application have at least the following technical effects: First, a multi-source sensor is used to perform multi-source perception fusion on the charging object of a mobile charging and storage device to obtain a target perception sequence. Next, a multi-level positioning of the charging object is performed based on the target perception sequence, obtaining a first target positioning feature and a second target positioning feature. Furthermore, perceptual interference correction is performed on the first and second target positioning features based on the multi-source sensor to obtain a third and fourth target positioning feature. Next, a driving hierarchy analysis is performed on the mobile charging and storage device based on the third and fourth target positioning features to construct a driving hierarchy factor. Then, driving control decisions are made for the mobile charging and storage device based on the driving hierarchy factor to obtain a first driving control space. This first driving control space is then iteratively optimized using a driving risk analysis model to determine a driving control strategy. Finally, driving management of the mobile charging and storage device is performed based on the driving control strategy, and driving control is updated in conjunction with the multi-source sensor. This method solves the technical problem of inaccurate target positioning in autonomous mobile charging and storage devices due to a single perception method in the prior art, achieving the technical effect of improving target positioning accuracy through multi-source data fusion and interference correction.

[0052] Embodiment 2 is based on the same inventive concept as the autonomous driving target positioning method based on multi-source data fusion in the above embodiment. Figure 2 As shown, the present application provides an autonomous driving target positioning system based on multi-source data fusion, wherein the system includes: The multi-source perception fusion module 11 performs multi-source perception fusion on the object to be charged of the mobile charging and storage device through a multi-source sensor to obtain a target perception sequence; the multi-level positioning module 12 performs multi-level positioning on the object to be charged according to the target perception sequence to obtain a first target positioning feature and a second target positioning feature; the perception interference correction module 13 performs perception interference correction on the first target positioning feature and the second target positioning feature based on the multi-source sensor to obtain a third target positioning feature and a fourth target positioning feature; the driving level analysis module 14 performs driving level analysis on the mobile charging and storage device according to the third target positioning feature and the fourth target positioning feature to construct a driving level factor; the driving control decision module 15 performs driving control decision on the mobile charging and storage device based on the driving level factor to obtain a first driving control space, and iteratively optimizes the first driving control space according to the driving risk analysis model to determine a driving control strategy; the driving management module 16 performs driving management on the mobile charging and storage device based on the driving control strategy, and updates the driving control in combination with the multi-source sensor.

[0053] Furthermore, the perceived interference correction module 13 is configured to perform the following method: Based on the target perception sequence, the multi-source sensor is subjected to perception scene collection to obtain multiple perception scene data; interference factors are identified on the multi-source sensor according to the multiple perception scene data to obtain a perception interference identification result; positioning error correction is performed on the first target positioning feature according to the perception interference identification result to obtain the third target positioning feature; positioning error correction is performed on the second target positioning feature according to the perception interference identification result to obtain the fourth target positioning feature.

[0054] Furthermore, the driving level parsing module 14 is configured to execute the following method: A target three-dimensional space is constructed based on the three-dimensional structural characteristics and spatial characteristics of the object to be charged; based on the target three-dimensional space, the navigation destination tolerance fitting is performed on the mobile storage and charging device according to the third feature of the target positioning to obtain the destination tolerance domain; the destination tolerance domain is corrected for reachability according to the device structural characteristics of the mobile storage and charging device to generate a destination tolerance position; the device geographic location of the mobile storage and charging device to the destination tolerance position is used as the first driving factor; the device external charging position corresponding to the destination tolerance position to the fourth feature of the target positioning is used as the second driving factor, and the first driving factor and the second driving factor are added to the driving hierarchy factor.

[0055] Furthermore, the driving control decision module 15 is configured to execute the following method: Driving scene features are collected based on the first driving factor and the second driving factor to construct a first driving scene model; obstacle features are labeled based on the first driving scene model to obtain a second driving scene model; multimodal driving control constraint information of the mobile charging and storage device is collected to generate driving control constraint conditions; based on the driving control constraint conditions, a driving control decision is made for the first driving factor according to the second driving scene model to obtain a first driving control decision set; based on the driving control constraint conditions, a driving control decision is made for the second driving factor according to the second driving scene model to obtain a second driving control decision set; and the first driving control decision set and the second driving control decision set are randomly combined to generate the first driving control space.

[0056] Furthermore, the driving control decision module 15 is configured to execute the following method: Constraints are configured based on multidimensional driving risk evaluation indicators of the driving risk analysis model to generate a driving risk constraint matrix, where the multidimensional driving risk evaluation indicators include driving safety risk, docking abnormality risk, and driving failure risk; based on the driving risk analysis model, driving risk constraint optimization is performed on the first driving control space according to the driving risk constraint matrix to establish a second driving control space; weights are configured based on the multidimensional driving risk evaluation indicators to obtain a comprehensive driving risk analysis model; comprehensive driving risk ranking optimization is performed on the second driving control space according to the comprehensive driving risk analysis model to obtain K driving control candidate solutions, where K is a positive integer; based on the driving risk analysis model and the comprehensive driving risk analysis model, variation optimization is performed on the K driving control candidate solutions to establish a third driving control space; and driving time minimization optimization is performed on the third driving control space to generate the driving control strategy.

[0057] Furthermore, the driving control decision module 15 is configured to execute the following method: A Wth driving control scheme is extracted based on the first driving control space, where W is a positive integer; a Wth simulated driving data is obtained by simulated driving of the mobile charging and storage device based on the Wth driving control scheme; the Wth simulated driving data is input into the driving risk analysis model to obtain a Wth driving risk matrix; if the Wth driving risk matrix satisfies the driving risk constraint matrix, the Wth driving control scheme is added to the second driving control space; and a driving risk constraint optimization is continued for the first driving control space based on the driving risk analysis model and the driving risk constraint matrix to obtain the second driving control space.

[0058] Furthermore, the driving control decision module 15 is configured to execute the following method: The second driving control space is subjected to difference identification based on the K driving control candidate solutions to obtain a plurality of driving control difference vectors; based on the driving hierarchy factor, the second driving control space is subjected to cross-mutation based on the plurality of driving control difference vectors to obtain a first driving variation group; based on the driving risk analytical model, the first driving variation group is subjected to driving risk constraint optimization according to the driving risk constraint matrix to establish a second driving variation group; the second driving variation group is expanded based on the K driving control candidate solutions to obtain a third driving variation group; and the third driving variation group is subjected to driving comprehensive risk ranking optimization according to the driving comprehensive risk analytical model to obtain the third driving control space.

[0059] Furthermore, the multi-source perception fusion module 11 is configured to perform the following method: The object to be charged is perceived in real time according to the multi-source sensor to obtain a first target perception set; the first target perception set is cleaned to obtain a second target perception set; and the second target perception set is integrated to generate the target perception sequence.

[0060] Furthermore, the multi-level positioning module 12 is used to perform the following method: The first target positioning feature includes the geographical location of the object to be charged, and the second target positioning feature includes the location of the charging interface of the object to be charged.

[0061] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0062] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0063] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. An autonomous driving target positioning method based on multi-source data fusion, characterized in that: The method comprises: Through the multi-source sensor, multi-source perception fusion is performed on the charging object of the mobile storage and charging device to obtain the target perception sequence; Perform multi-level positioning of the object to be charged according to the target sensing sequence to obtain a first target positioning feature and a second target positioning feature; Performing perceptual interference correction on the first target positioning feature and the second target positioning feature based on the multi-source sensor to obtain a third target positioning feature and a fourth target positioning feature; Performing a driving level analysis on the mobile storage and charging device according to the third target positioning feature and the fourth target positioning feature to construct a driving level factor; Performing a driving control decision on the mobile charging and storage device based on the driving hierarchy factor to obtain a first driving control space, and iteratively optimizing the first driving control space according to a driving risk analytical model to determine a driving control strategy; The mobile storage and charging device is driven and managed based on the driving control strategy, and the driving control is updated in combination with the multi-source sensor.

2. The autonomous driving target positioning method based on multi-source data fusion according to claim 1, characterized in that: Performing perceptual interference correction on the first target positioning feature and the second target positioning feature based on the multi-source sensor to obtain a third target positioning feature and a fourth target positioning feature, including: Based on the target perception sequence, the multi-source sensor performs perception scene acquisition to obtain a plurality of perception scene data; Identifying interference factors on the multi-source sensor according to the plurality of perception scene data to obtain a perception interference identification result; Performing positioning error correction on the first target positioning feature according to the perception interference recognition result to obtain the third target positioning feature; A positioning error correction is performed on the second target positioning feature according to the perception interference recognition result to obtain the fourth target positioning feature.

3. The autonomous driving target positioning method based on multi-source data fusion according to claim 1, characterized in that: The driving level analysis of the mobile storage and charging device is performed according to the third feature of the target positioning and the fourth feature of the target positioning, and a driving level factor is constructed, including: Constructing a target three-dimensional space according to the three-dimensional structural characteristics and spatial characteristics of the object to be charged; Based on the target three-dimensional space, performing navigation endpoint tolerance fitting on the mobile storage and charging device according to the third target positioning feature to obtain an endpoint tolerance bit domain; Performing reachability correction on the endpoint tolerance bit field according to the device structural characteristics of the mobile storage and charging device to generate an endpoint tolerance position; The first driving factor is the distance from the geographical location of the mobile charging device to the tolerance location of the terminal; The fourth feature of the target positioning is determined by taking the device external charging position corresponding to the endpoint tolerance position as the second driving factor, and adding the first driving factor and the second driving factor to the driving hierarchy factor.

4. The autonomous driving target positioning method based on multi-source data fusion according to claim 1, characterized in that: Performing a driving control decision on the mobile charging and storage device based on the driving hierarchy factor to obtain a first driving control space includes: collecting driving scene features based on the first driving factor and the second driving factor, and constructing a first driving scene model; Obstacle feature annotation is performed based on the first driving scene model to obtain a second driving scene model; Collecting multimodal driving control constraint information of the mobile storage and charging device to generate driving control constraint conditions; Based on the driving control constraint condition, performing a driving control decision on the first driving factor according to the second driving scenario model to obtain a first driving control decision set; Based on the driving control constraint condition, performing a driving control decision on the second driving factor according to the second driving scenario model to obtain a second driving control decision set; The first driving control space is generated by randomly combining the first driving control decision set and the second driving control decision set.

5. The autonomous driving target positioning method based on multi-source data fusion according to claim 1, characterized in that: Iteratively optimizing the first driving control space according to the driving risk analytical model to determine a driving control strategy includes: Performing constraint configuration based on the multidimensional driving risk evaluation indicators of the driving risk analysis model to generate a driving risk constraint matrix, wherein the multidimensional driving risk evaluation indicators include driving safety risk, docking abnormality risk, and driving failure risk; Based on the driving risk analytical model, performing driving risk constraint optimization on the first driving control space according to the driving risk constraint matrix to establish a second driving control space; Based on the multi-dimensional driving risk evaluation indicators, weight configuration is performed to obtain a comprehensive driving risk analysis model; Performing driving comprehensive risk sorting and optimization on the second driving control space according to the driving comprehensive risk analytical model to obtain K driving control candidate solutions, where K is a positive integer; Based on the driving risk analytical model and the driving comprehensive risk analytical model, performing mutation optimization on the K driving control candidate solutions to establish a third driving control space; The driving time is minimized and optimized according to the third driving control space to generate the driving control strategy.

6. The autonomous driving target positioning method based on multi-source data fusion according to claim 5, characterized in that: Based on the driving risk analytical model, performing driving risk constraint optimization on the first driving control space according to the driving risk constraint matrix to establish a second driving control space includes: Extracting a Wth driving control scheme according to the first driving control space, where W is a positive integer; Performing simulated driving on the mobile charging device based on the Wth driving control scheme to obtain Wth simulated driving data; inputting the Wth simulated driving data into the driving risk analysis model to obtain a Wth driving risk matrix; If the Wth driving risk matrix satisfies the driving risk constraint matrix, adding the Wth driving control scheme to the second driving control space; The first driving control space is further optimized according to the driving risk analytical model and the driving risk constraint matrix to obtain the second driving control space.

7. The autonomous driving target positioning method based on multi-source data fusion according to claim 5, characterized in that: Based on the driving risk analysis model and the driving comprehensive risk analysis model, performing mutation optimization on the K driving control candidate solutions to establish a third driving control space, including: performing difference identification on the second driving control space according to the K driving control candidate solutions to obtain a plurality of driving control difference vectors; Based on the driving level factor, performing cross mutation on the second driving control space according to the plurality of driving control difference vectors to obtain a first group of driving mutations; Based on the driving risk analysis model, performing driving risk constraint optimization on the first group of driving variations according to the driving risk constraint matrix to establish a second group of driving variations; Expanding the second group of driving variations according to the K candidate driving control schemes to obtain a third group of driving variations; The third driving control space is obtained by performing driving comprehensive risk ranking optimization on the third group of driving variations according to the driving comprehensive risk analytical model.

8. The autonomous driving target positioning method based on multi-source data fusion according to claim 1, characterized in that: The multi-source sensor performs multi-source perception fusion on the charging object of the mobile charging and storage device to obtain the target perception sequence, including: Performing real-time perception of the object to be charged according to the multi-source sensor to obtain a first set of target perception; cleaning the first target perception set to obtain a second target perception set; The second set of target perception is fused to generate the target perception sequence.

9. The autonomous driving target positioning method based on multi-source data fusion according to claim 1, characterized in that: The first target positioning feature includes the geographical location of the object to be charged, and the second target positioning feature includes the location of the charging interface of the object to be charged.

10. The autonomous driving target positioning system based on multi-source data fusion is characterized by: A system for implementing the autonomous driving target positioning method based on multi-source data fusion according to any one of claims 1 to 9, comprising: Multi-source perception fusion module: This module uses multi-source sensors to perform multi-source perception fusion on the charging object of the mobile charging and storage device to obtain the target perception sequence; Multi-level positioning module: performing multi-level positioning of the object to be charged according to the target sensing sequence, and obtaining a first target positioning feature and a second target positioning feature; A perception interference correction module is configured to perform perception interference correction on the first target positioning feature and the second target positioning feature based on the multi-source sensor to obtain a third target positioning feature and a fourth target positioning feature; A driving level analysis module: performing driving level analysis on the mobile storage and charging device according to the third feature of target positioning and the fourth feature of target positioning, and constructing a driving level factor; A driving control decision module: performs driving control decision-making on the mobile charging and storage device based on the driving hierarchical factors to obtain a first driving control space, and iteratively optimizes the first driving control space according to a driving risk analytical model to determine a driving control strategy; Driving management module: performs driving management on the mobile charging and storage device based on the driving control strategy, and updates driving control in combination with the multi-source sensor.

Images