A path tracking system and method based on laser radar
Through a path tracking system based on lidar, it quickly integrates and processes three-dimensional point cloud data, builds a virtual model and simulates customer mobile paths, solves the problem of long point cloud registration time in large shopping malls, realizes intelligent path guidance, and improves navigation Efficiency and convenience.
Patent Information
- Application Number
- CN202311718586.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-14
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2043-12-14
AI Technical Summary
In large shopping malls, due to the large scale of point cloud data and the long registration time of point cloud, the distribution of personnel within the shopping mall cannot be obtained in time, and it is impossible to provide appropriate path guidance services for customers who have entered the shopping mall.
Design a path tracking system based on lidar, including image acquisition module, data storage module, point cloud registration module, virtual reality module, path planning module and interaction module, and obtain three-dimensional point cloud data through multiple lidars to perform data Fusion and preprocessing, build a virtual model of the mall, simulate customer movement paths, judge channel congestion, and recommend customer movement paths through the shortest path algorithm.
It quickly obtains the distribution of personnel within the mall, provides intelligent path guidance, improves the efficiency and convenience of customer navigation, and optimizes the guidance effect in the mall.
Smart Images

Figure CN117784159B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of path guidance technology, and in particular to a laser radar-based path tracking system and method. Background Art
[0002] 3D LiDAR is a device that performs 3D scanning based on LiDAR technology. It is usually used in surveying and mapping, unmanned driving, drones, security monitoring, robot navigation, intelligent driving and other related fields. It scans the target object by emitting lasers, and then uses the time it takes for the received laser to be reflected back to determine the distance of the object, thereby forming point cloud data; point cloud registration refers to the process of aligning or matching two or more point cloud data. In large shopping malls, the 3D point cloud of the entire shopping mall can be generated by fusing the local point clouds generated by multiple LiDARs; however, the point cloud registration time will be affected by the scale of the point cloud data. In large shopping malls, it is necessary to process the point cloud data of multiple LiDARs, and the amount of data is very large. The point cloud registration takes a long time, resulting in the inability to obtain the overall distribution of personnel in the mall in a timely manner, and it is impossible to provide appropriate path guidance services for new customers entering the mall; for this reason, how to obtain the overall distribution of personnel in the mall in a timely manner and provide intelligent guidance to customers has become a problem that needs to be solved. Summary of the invention
[0003] The purpose of the present invention is to provide a path tracking system and method based on laser radar to solve the problems raised in the above background technology.
[0004] In one aspect of the present invention, a path tracking system based on laser radar is provided, comprising: an image acquisition module, a data storage module, a point cloud registration module, a virtual reality module, a path planning module and an interaction module; the output end of the image acquisition module is interconnected with the input end of the point cloud registration module, and the three-dimensional point cloud data in the shopping mall is obtained through multiple laser radars; the output end of the data storage module is interconnected with the input end of the virtual reality module, and is used to store the historical movement path data of customers in the shopping mall and the pedestrian flow data of the shopping mall; the output end of the point cloud registration module is interconnected with the input end of the data storage module, and is used to register the three-dimensional point cloud of the laser radar to generate a complete three-dimensional point cloud of the shopping mall; the output end of the virtual reality module is interconnected with the input end of the path planning module, and is used to simulate the movement path of the customer and obtain the road condition information in the shopping mall; the output end of the path planning module is interconnected with the input end of the interaction module, and a movement path is recommended for the customer based on the road condition information in the shopping mall; the interaction module is deployed at the entrance of the shopping mall to provide the customer with movement path information and estimated arrival time.
[0005] Specifically, the point cloud registration module also includes: a preprocessing unit and a point cloud registration unit; the preprocessing unit is used to preprocess the point cloud data acquired by the laser radar; the point cloud registration unit is used to align and fuse the point cloud data from different laser radars to generate an overall three-dimensional point cloud within the shopping mall.
[0006] Specifically, the virtual reality module constructs a virtual model of the store based on the complete three-dimensional point cloud information of the shopping mall, and applies it to the virtual reality environment; at the entrance of the shopping mall, a virtual model of the customer and the customer's movement path are generated at a speed consistent with the flow of people, and the traffic conditions in the shopping mall for a whole day are simulated to obtain the traffic conditions in the shopping mall at different times.
[0007] The path planning module also includes a judgment unit, a weighted graph generation unit and a planning unit; the judgment unit judges whether the passages in the mall are congested according to the time when the customer enters the mall and the simulated traffic conditions in the mall; the weighted graph generation unit is used to construct a graph model in the mall, with vertices representing different stores in the mall, edges representing passages connecting different stores, the travel time corresponding to the edge in the passage is the weight of the edge, and the edge weight is positive infinity when the passage is congested; the planning unit determines a moving path with the shortest estimated travel time for the customer through a shortest path algorithm.
[0008] In another aspect of the present invention, a laser radar-based path tracking method is provided, comprising the following steps:
[0009] S5-1, obtain three-dimensional point cloud data from the laser radar and obtain three-dimensional coordinate information of the entire shopping mall;
[0010] S5-2, analyzing the historical three-dimensional coordinate information of customers in the mall and determining the customer's movement trajectory model;
[0011] S5-3, using the three-dimensional coordinate information, constructing a virtual model of the shopping mall, and applying it to the virtual reality environment, using the customer's movement trajectory model to simulate the real environment, and obtaining the traffic information in the shopping mall;
[0012] S5-4, based on the traffic information in the mall, recommend the best route to the destination for the customer.
[0013] Specifically, in step S5-1, the acquisition of the three-dimensional coordinate information of the entire shopping mall further includes the following steps:
[0014] Preprocess the three-dimensional point cloud data obtained by each laser radar;
[0015] Align and fuse point cloud data from different lidars to generate a 3D point cloud for the entire shopping mall.
[0016] Specifically, in step S5-2, determining the customer's action trajectory model further includes the following steps:
[0017] For each customer entering from each mall entrance, the customer's movement trajectory is represented by a tree diagram; based on the three-dimensional point cloud of the entire mall, the customer's complete path from entering the mall to leaving the mall is obtained, starting from the mall entrance, and each time a store is passed, a node is generated in the tree diagram to represent the store, and the node and the store are in a many-to-one relationship; the branches of the node represent the path between stores; different nodes are distinguished by different numbers, and the nodes are related to the customer's movement path. For a customer, there is no node with the same number in the tree diagram of his movement path; for different customers, starting from the same starting point, when there is the same movement path, the nodes on the same movement path have the same number, and between different customers, at most the nodes on a section of the path have the same number;
[0018] The tree diagrams of all customer movement paths entering the mall from the same entrance are merged into a whole, and the nodes with the same number are merged. The nodes are also used to record the number of people. Each time the nodes with the same number are merged, the number of people corresponding to the node is increased by one. K complete tree diagrams are obtained, where k is the number of mall entrances.
[0019] The node number includes the customer's historical movement path information and the customer's location information. The customer's location information and the customer's movement path can be used to predict the customer's next direction, that is, which node the customer will go to next. The node number and the store in the mall have a many-to-one relationship. Multiple nodes with different numbers can correspond to the same store. When customers take different movement paths to reach the same store, the node the customer will go to next will be affected by the customer's previous movement path.
[0020] Specifically, in step S5-3, simulating the real environment using the customer's motion trajectory model to obtain the traffic information in the mall includes the following steps:
[0021] S8-1, A(0,1) represents the entry node, A(1,1), A(1,2), ...A(1,n) represent the next level nodes of the entry node, and NA(0,1), NA(1,1), NA(1,2), ...NA(1,n) represent the number of people corresponding to nodes A(0,1), A(1,1), A(1,2), ...A(1,n), n is the number of stores that can be reached directly from the entrance without passing through other stores;
[0022] S8-2, calculate the probability PA(1,1), PA(1,2), ...PA(1,n) of the virtual character going to the next level node A(1,1), A(1,2), ...A(1,n), PA(1,1) = NA(1,1) / NA(0,1), PA(1,2) = NA(1,2) / NA(0,1), ..., PA(1,n) = NA(1,n) / NA(0,1), and go to the next level node according to the probability through random numbers;
[0023] S8-3, at the current node, continue to go to the next level node according to probability through random numbers; repeat step S8-3 until the virtual character reaches the mall exit, and obtain a complete movement path of the virtual character;
[0024] S8-4, based on the complete movement paths of all virtual characters and the average movement speed of customers in the mall, obtain the number of people in each channel at different times in the mall;
[0025] By determining the nodes reached by the virtual character one by one, the entire movement path of the virtual character in the mall is obtained; starting from the entrance node, the historical data includes the number of customers who went from the entrance node to different nodes, and the probability of customers going from the entrance node to different nodes is determined according to the ratio of the number of people going from the entrance node to the total number of people at the entrance node, wherein the total number of people at the entrance node is recorded on the node number, and the number of people going from the entrance node to different nodes is recorded on the node number of the next level node; similarly, when the virtual character arrives at the next level node, the number of people recorded on the node is continued to be judged until the virtual character arrives at the mall exit; the movement paths of the virtual characters are all from historical customers, and there will be no situation where the virtual character cannot leave the mall and stays in the mall;
[0026] In the virtual environment, the production speed of virtual characters is determined according to the flow of people at the mall entrance. Production of virtual characters starts from the opening time of the mall and stops after the mall closes, thereby simulating the distribution of customers in the mall throughout the day.
[0027] Specifically, in step S5-4, the step of recommending the best route to the destination for the customer further includes the following steps:
[0028] S9-1, with vertices representing different stores in the mall, edges representing channels connecting different stores, and the travel time of the channels corresponding to the edges as the weight of the edges, the initial graph model of the mall is obtained;
[0029] S9-2, adjusting the weight of the initial graph model in the mall; judging whether the channel will be congested according to the design capacity of the channel in the mall and the number of customers in the channel when entering the mall; if the channel will be congested, setting the weight of the edge corresponding to the congested channel to positive infinity, and obtaining the graph model of the mall when the customer enters the mall;
[0030] The present invention is used to provide customers with the shortest path guidance, so the weight in the graph model is the travel time, which can be determined based on the historical movement path information of customers in the mall. The estimated travel time of the passages connecting different stores can be obtained through the time difference of customers arriving at two stores in the historical data. When the passage is congested, it will greatly affect the customer experience, so the weight of the congested passage is set to positive infinity.
[0031] On the graph model in the mall, the Dijkstra algorithm is used to recommend the best movement path for customers, including the following steps:
[0032] S10-1, in the graph model in the mall, specify the mall entrance as the starting point, introduce arrays U and V, U is used to record the vertices for which the shortest path has not been found and the path length from the vertex to the starting point. If the vertex is not adjacent to the starting point, the path length is infinite; V is used to record the vertices for which the shortest path has been found and the shortest path length from the vertex to the starting point; the path length refers to the sum of the weights of all edges included in the path;
[0033] S10-2, the array V initially includes only the starting point, and the data U includes all vertices except the starting point.
[0034] S10-3, find the vertex with the shortest path from array U and add it to array V; at the same time, remove the vertex with the shortest path from array U and update the distance from each vertex in array U to the starting point;
[0035] S10-4, repeat step S10-3 until all vertices are traversed;
[0036] S10-5, find the shortest path in the array V according to the vertex corresponding to the destination that the customer wants to go to.
[0037] In particular, there is no need to wait until customers enter the mall before the Dijkstra algorithm recommends the best movement path for customers; since the distribution of people in the mall throughout the day has been obtained in the virtual environment, the Dijkstra algorithm can be used in advance to calculate the shortest path from each entrance to a high-popularity store during the peak period based on the simulated distribution of people during the peak period, further speeding up the speed of recommending the path.
[0038] Compared with the prior art, the beneficial effects achieved by the present invention are: by utilizing multiple laser radars, reasonable deployment and data fusion algorithms, the three-dimensional coordinate information of the store space can be acquired; by utilizing the three-dimensional coordinate information, a navigation system can be developed to provide customers with positioning and intelligent guidance services, and customers can use mobile phones or interactive modules in the mall to quickly find the appropriate path to the target store in the mall, thereby improving efficiency and convenience; by utilizing virtual reality technology combined with three-dimensional coordinate information, the movement path of customers in the mall can be simulated, which helps to optimize the guidance effect in the mall. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0040] Figure 1 is a schematic structural diagram of a laser radar-based path tracking system according to an embodiment of the present invention;
[0041] Figure 2 It is a schematic diagram of a structural diagram of the distribution of stores in a shopping mall according to an embodiment of the present invention;
[0042] Figure 3 It is a partial tree diagram of a shopping mall in an embodiment of the present invention. DETAILED DESCRIPTION
[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0044] In the embodiments of the present invention, please refer to Figure 1, a path tracking system based on laser radar is provided, comprising: an image acquisition module, a data storage module, a point cloud registration module, a virtual reality module, a path planning module and an interactive module; the output end of the image acquisition module is interconnected with the input end of the point cloud registration module, and the three-dimensional point cloud data in the shopping mall is obtained through multiple laser radars; the output end of the data storage module is interconnected with the input end of the virtual reality module, and is used to store the historical movement path data of customers in the shopping mall and the pedestrian flow data of the shopping mall; the output end of the point cloud registration module is interconnected with the input end of the data storage module, and is used to register the three-dimensional point cloud of the laser radar to generate a complete three-dimensional point cloud of the shopping mall; the output end of the virtual reality module is interconnected with the input end of the path planning module, and is used to simulate the movement path of the customer and obtain the road condition information in the shopping mall; the output end of the path planning module is interconnected with the input end of the interactive module, and a movement path is recommended for the customer based on the road condition information in the shopping mall; the interactive module is deployed at the entrance of the shopping mall to provide the customer with movement path information and estimated arrival time.
[0045] The point cloud registration module also includes: a preprocessing unit and a point cloud registration unit; the preprocessing unit is used to preprocess the three-dimensional point cloud data acquired by the laser radar; the point cloud registration unit is used to align and fuse the point cloud data from different laser radars to generate an overall three-dimensional point cloud in the shopping mall.
[0046] The virtual reality module constructs a virtual model of the store based on the complete three-dimensional point cloud information of the shopping mall and applies it to the virtual reality environment; at the entrance of the shopping mall, a virtual model of the customer and the customer's movement path are generated at a speed consistent with the flow of people, and the traffic conditions in the shopping mall for a whole day are simulated to obtain the traffic conditions in the shopping mall at different times.
[0047] The path planning module also includes a judgment unit, a weighted graph generation unit and a planning unit; the judgment unit judges whether the passages in the mall are congested according to the time when the customer enters the mall and the simulated traffic conditions in the mall; the weighted graph generation unit is used to construct a graph model in the mall, with vertices representing different stores in the mall, edges representing passages connecting different stores, the travel time corresponding to the edge in the passage is the weight of the edge, and the edge weight is positive infinity when the passage is congested; the planning unit determines a moving path with the shortest estimated travel time for the customer through a shortest path algorithm.
[0048] In an embodiment of the present invention, a laser radar-based path tracking method is provided, comprising the following steps:
[0049] S5-1, obtaining three-dimensional point cloud data from the laser radar to obtain three-dimensional coordinate information of the entire shopping mall; including the following steps:
[0050] Preprocess the three-dimensional point cloud data obtained by each laser radar;
[0051] Align and fuse point cloud data from different LiDARs to generate a 3D point cloud of the entire mall;
[0052] S5-2, analyzing the historical three-dimensional coordinate information of customers in the mall and determining the customer's movement trajectory model;
[0053] For each customer entering from each mall entrance, the customer's movement trajectory is represented by a tree diagram; based on the three-dimensional unit of the entire mall, the customer's complete path from entering the mall to leaving the mall is obtained, starting from the mall entrance, and each time a store is passed, a node is generated in the tree diagram to represent the store, and the node and the store are in a many-to-one relationship; the branches of the node represent the path between stores; different nodes are distinguished by different numbers, and the nodes are related to the customer's movement path. For a customer, there is no node with the same number in the tree diagram of his movement path; for different customers, starting from the same starting point, when there is the same movement path, the nodes on the same movement path have the same number, and between different customers, the node numbers on at most one section of the path are the same;
[0054] The tree diagrams of all customer movement paths entering the mall from the same entrance are merged into a whole, and the nodes with the same number are merged. The nodes are also used to record the number of people. Each time the nodes with the same number are merged, the number of people corresponding to the node is increased by one. K complete tree diagrams are obtained, where k is the number of mall entrances.
[0055] S5-3, using the three-dimensional coordinate information, constructing a virtual model of the shopping mall, and applying it to the virtual reality environment, using the customer's movement trajectory model to simulate the real environment, and obtaining the traffic information in the shopping mall;
[0056] Obtain the historical traffic data of each entrance of the shopping mall, and take the average value as the virtual traffic data of each entrance in the virtual reality environment; based on the virtual traffic data of each entrance, generate a virtual character at each entrance at a speed that matches the virtual traffic data; simulate the movement path of each virtual character according to the complete tree diagram, including the following steps:
[0057] S8-1, A(0,1) represents the entry node, A(1,1), A(1,2), ...A(1,n) represent the next level nodes of the entry node, and NA(0,1), NA(1,1), NA(1,2), ...NA(1,n) represent the number of people corresponding to nodes A(0,1), A(1,1), A(1,2), ...A(1,n), n is the number of stores that can be reached directly from the entrance without passing through other stores;
[0058] S8-2, calculate the probability PA(1,1), PA(1,2), ...PA(1,n) of the virtual character going to the next level node A(1,1), A(1,2), ...A(1,n), PA(1,1) = NA(1,1) / NA(0,1), PA(1,2) = NA(1,2) / NA(0,1), ..., PA(1,n) = NA(1,n) / NA(0,1), and go to the next level node according to the probability through random numbers;
[0059] S8-3, at the current node, continue to go to the next level node according to probability through random numbers; repeat step S8-3 until the virtual character reaches the mall exit, and obtain a complete movement path of the virtual character;
[0060] S8-4, based on the complete movement paths of all virtual characters and the average movement speed of customers in the mall, obtain the number of people in each channel at different times in the mall;
[0061] S5-4, based on the traffic information in the mall, recommend the best route to the destination for the customer. The specific steps are as follows:
[0062] The vertices represent different stores in the mall, the edges represent the channels connecting different stores, and the travel time of the channels corresponding to the edges is the weight of the edges, thus obtaining the initial graph model of the mall;
[0063] The weight of the initial graph model in the mall is adjusted; based on the design capacity of the aisle in the mall and the number of customers in the aisle when entering the mall, it is determined whether the aisle will be congested; if the aisle is congested, the weight of the edge corresponding to the congested aisle is set to positive infinity, and the graph model of the mall when customers enter the mall is obtained;
[0064] According to the graph model in the mall, the Dijkstra algorithm is used to recommend the best movement path for customers:
[0065] S10-1, in the graph model in the mall, specify the mall entrance as the starting point, introduce arrays U and V, U is used to record the vertices for which the shortest path has not been found and the path length from the vertex to the starting point. If the vertex is not adjacent to the starting point, the path length is infinite; V is used to record the vertices for which the shortest path has been found and the shortest path length from the vertex to the starting point; the path length refers to the sum of the weights of all edges included in the path;
[0066] S10-2, the array V initially includes only the starting point, and the data U includes all vertices except the starting point.
[0067] S10-3, find the vertex with the shortest path from array U and add it to array V; at the same time, remove the vertex with the shortest path from array U and update the distance from each vertex in array U to the starting point;
[0068] S10-4, repeat step S10-3 until all vertices are traversed;
[0069] S10-5, find the shortest path in the array V according to the vertex corresponding to the destination that the customer wants to go to.
[0070] See also Figure 2 , providing a structural diagram of the store distribution in a shopping mall; serial number 1 represents the entrance and exit of the shopping mall, serial numbers 2 to 9 represent the stores in the shopping mall, and when the number of customers moving along the following three moving paths 1258741, 12521 and 1479631 is 10, 20, 10, please refer to Figure 3 , providing a partial tree diagram in the mall; in the figure, number A(0,1) corresponds to the entrance of the mall, A(6,1), A(4,2) and A(6,2) correspond to the exit of the mall, A(2,1) and A(3,2) correspond to the store numbered 2, A(5,1) and A(1,2) correspond to the store numbered 4, A(2,1) corresponds to the store numbered 5, A(4,3) corresponds to the store numbered 6, A(4,1) and A(2,3) correspond to the store numbered 7, A(3,1) corresponds to the store numbered 8, and A(3,3) corresponds to the store numbered 8; when the flow of people at the mall entrance is 10 people per minute, a virtual character is generated at the mall entrance every 6 seconds, and the movement path of the virtual character is planned; according to the formation process of the tree diagram, the number of people corresponding to NA(0,1), NA(1,1) and NA(1,2) are respectively 40, 30 and 10, so the virtual character has a 30 / 40=75% probability of going to node A(1,1), that is, the store with serial number 2, and a 25% probability of going to node A(1,2), that is, the store with serial number 4; if the virtual character goes to node A(1,2), there is no branch in the subsequent tree diagram, and the movement path of the virtual character is completely determined; if the virtual character goes to node A(1,2), according to the formation process of the tree diagram, the number of people corresponding to NA(2,1), NA(3,1) and NA(3,2) are 30, 10 and 20 respectively, and the virtual character has a 10 / 30=33% probability of going to node A(3,1), that is, the store with serial number 8, and a 67% probability of going to node A(3,2), that is, the store with serial number 2; the next node that the virtual character needs to go to is continuously determined by generating random numbers until the virtual character reaches the exit of the mall;
[0071] Since the node number records the historical movement path information of the virtual character, even if nodes A(2,1) and A(3,2) both correspond to the store with sequence number 2, the nodes that the virtual character will go to next are completely different;
[0072] After determining the movement path of the virtual character, the location of the virtual character at different times can be determined according to the average time spent by customers in different stores in historical data; the distribution of all customers in the mall can be obtained according to the location of all virtual characters at different times;
[0073] Based on the initial graph model in the mall, whether the channel will be congested is determined according to the design capacity of the channel and the number of people in the channel; if the channel is congested, the weight of the edge corresponding to the congested channel is set to positive infinity, otherwise the weight of the edge corresponding to the channel remains unchanged; the best movement path can be recommended to customers through the Dijkstra algorithm.
[0074] By accelerating simulation in a virtual environment, the distribution of people in a shopping mall throughout the day can be obtained. Therefore, based on the distribution of people during peak hours obtained through simulation, the Dijkstra algorithm can be used in advance to calculate the shortest paths from various entrances to high-popularity stores during peak hours, further speeding up the speed of recommending paths without having to wait until customers enter the mall to provide intelligent guidance to customers.
[0075] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0076] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A path tracking method based on laser radar, characterized in that: The following steps are involved: S5-1, obtain three-dimensional point cloud data from the laser radar and obtain three-dimensional coordinate information of the entire shopping mall; Preprocess the three-dimensional point cloud data obtained by each laser radar; Align and fuse point cloud data from different LiDARs to generate a 3D point cloud of the entire mall; S5-2, analyzing the historical three-dimensional coordinate information of customers in the mall and determining the customer's movement trajectory model; For each customer entering from each mall entrance, a tree diagram is used to represent the customer's movement trajectory; based on the 3D point cloud of the entire mall, the customer's complete path from entering the mall to leaving the mall is obtained. Starting from the mall entrance, each time a store is passed, a node is generated in the tree diagram to represent the store, and the node and the store are in a many-to-one relationship; The branches of the nodes represent the paths between stores. Different nodes are distinguished by different numbers. The nodes are related to the customer's movement path. For a customer, there are no nodes with the same number in the tree diagram of his movement path. For different customers, starting from the same starting point, when there is the same movement path, the nodes on the same movement path have the same number. For different customers, at most one node on a path has the same number. The tree diagrams of all customer movement paths entering the mall from the same entrance are merged into a whole, and the nodes with the same number are merged. The nodes are also used to record the number of people. Each time the nodes with the same number are merged, the number of people corresponding to the node is increased by one. K complete tree diagrams are obtained, where k is the number of mall entrances. S5-3, using the three-dimensional coordinate information, constructing a virtual model of the shopping mall, and applying it to the virtual reality environment, using the customer's movement trajectory model to simulate the real environment, and obtaining the traffic information in the shopping mall; Obtain the historical traffic data of each entrance of the shopping mall, and take the average value as the virtual traffic data of each entrance in the virtual reality environment; Based on the virtual human flow data of each entrance, a virtual character is generated at each entrance at a speed that matches the virtual human flow data; and the moving path of each virtual character is simulated according to the complete tree diagram, including the following steps: S8-1, A(0,1) represents the entry node, A(1,1), A(1,2), ...A(1,n) represent the next level nodes of the entry node, and NA(0,1), NA(1,1), NA(1,2), ...NA(1,n) represent the number of people corresponding to nodes A(0,1), A(1,1), A(1,2), ...A(1,n), n is the number of stores that can be reached directly from the entrance without passing through other stores; S8-2, calculate the probability PA(1,1), PA(1,2), ...PA(1,n) of the virtual character going to the next level node A(1,1), A(1,2), ...A(1,n), PA(1,1) = NA(1,1) / NA(0,1), PA(1,2) = NA(1,2) / NA(0,1), ..., PA(1,n) = NA(1,n) / NA(0,1), and go to the next level node according to the probability through random numbers; S8-3, at the current node, continue to go to the next level node according to probability through random numbers; repeat step S8-3 until the virtual character reaches the mall exit, and obtain a complete movement path of the virtual character; S8-4, based on the complete movement paths of all virtual characters and the average movement speed of customers in the mall, obtain the number of people in each channel at different times in the mall; S5-4, based on the traffic information in the mall, recommending the best route to the destination for the customer, including the following steps: S9-1, with vertices representing different stores in the mall, edges representing channels connecting different stores, and the travel time of the channels corresponding to the edges as the weight of the edges, the initial graph model of the mall is obtained; S9-2, adjusting the weight of the initial graph model in the mall; judging whether the channel will be congested according to the design capacity of the channel in the mall and the number of customers in the channel when entering the mall; if the channel will be congested, setting the weight of the edge corresponding to the congested channel to positive infinity, otherwise the weight of the edge corresponding to the channel remains unchanged, and obtaining the graph model of the mall when customers enter the mall; According to the destination that the customer wants to go to, the Dijkstra algorithm is used to recommend the best movement path for the customer, including the following steps: S10-1, in the graph model in the mall, specify the mall entrance as the starting point, introduce arrays U and V, U is used to record the vertices for which the shortest path has not been found and the path length from the vertex to the starting point. If the vertex is not adjacent to the starting point, the path length is infinite; V is used to record the vertices for which the shortest path has been found and the shortest path length from the vertex to the starting point; the path length refers to the sum of the weights of all edges included in the path; S10-2, the array V initially includes only the starting point, and the data U includes all vertices except the starting point. S10-3, find the vertex with the shortest path from array U and add it to array V; at the same time, remove the vertex with the shortest path from array U and update the distance from each vertex in array U to the starting point; S10-4, repeat step S10-3 until all vertices are traversed; S10-5, find the shortest path in the array V according to the vertex corresponding to the destination that the customer wants to go to.
2. A laser radar-based path tracking system, implementing a laser radar-based path tracking method as claimed in claim 1, characterized in that: It includes an image acquisition module, a data storage module, a point cloud registration module, a virtual reality module, a path planning module and an interactive module; the output end of the image acquisition module is interconnected with the input end of the point cloud registration module to obtain the three-dimensional point cloud data in the mall through the laser radar; the output end of the data storage module is interconnected with the input end of the virtual reality module to store the historical movement path data of customers in the mall and the pedestrian flow data of the mall; the output end of the point cloud registration module is interconnected with the input end of the data storage module to register the three-dimensional point cloud of the laser radar to generate a complete three-dimensional point cloud of the mall; the output end of the virtual reality module is interconnected with the input end of the path planning module to simulate the movement path of the customer and obtain the road condition information in the mall; the output end of the path planning module is interconnected with the input end of the interactive module to recommend a movement path for the customer based on the road condition information in the mall; the interactive module is deployed at the entrance of the mall to provide the customer with movement path information and estimated arrival time.
3. A laser radar-based path tracking system according to claim 2, characterized in that: The point cloud registration module also includes: a preprocessing unit and a point cloud registration unit; the preprocessing unit is used to preprocess the point cloud data acquired by the laser radar; the point cloud registration unit is used to align and fuse the point cloud data from different laser radars to generate an overall three-dimensional point cloud within the shopping mall.
4. A laser radar-based path tracking system according to claim 2, characterized in that: The virtual reality module constructs a virtual model of the store based on the complete three-dimensional point cloud information of the shopping mall and applies it to the virtual reality environment; at the entrance of the shopping mall, a virtual model of the customer and the customer's movement path are generated at a speed consistent with the flow of people, and the traffic conditions in the shopping mall for a whole day are simulated to obtain the traffic conditions in the shopping mall at different times.
5. The laser radar-based path tracking system according to claim 2, characterized in that: The path planning module also includes a judgment unit, a weighted graph generation unit and a planning unit; the judgment unit judges whether the passages in the mall are congested according to the time when the customer enters the mall and the simulated traffic conditions in the mall; the weighted graph generation unit is used to construct a graph model in the mall, with vertices representing different stores in the mall, edges representing passages connecting different stores, the travel time corresponding to the edge in the passage is the weight of the edge, and the edge weight is positive infinity when the passage is congested; the planning unit determines a moving path with the shortest estimated travel time for the customer through a shortest path algorithm.
