Port dynamic point cloud map updating method and system and storage medium

Abstract

The invention relates to a port dynamic point cloud map updating method and system and a storage medium, and the method comprises the steps: U1, enabling a vehicle to carry out a port operation task, obtaining the data information of an original global point cloud map of a port, obtaining the point cloud data information of a road in real time based on a vehicle-mounted laser radar, and obtaining the data information of the pose of the vehicle in real time based on a vehicle-mounted IMU, the method comprises the following steps: acquiring data information of an absolute pose of a vehicle in real time based on a vehicle-mounted inertial navigation sensor, extracting local features of a point cloud map by adopting a self-adaptive upsampling algorithm based on a chaotic mapping sequence, and constructing a real-time dense local point cloud map to obtain data information of the real-time dense local point cloud map. According to the method, environment change and map updating delay can be avoided, interference on current matching positioning is reduced, a reliable matching positioning result is output in real time, and use of the point cloud map on the storage space of the controller is reduced as much as possible.

Description

Technical Field

[0001] The present invention relates to the technical field of port point cloud maps, and in particular to a method, system and storage medium for updating a port dynamic point cloud map. Background Art

[0002] When unmanned operation vehicles operate in a closed campus environment, they usually require stable and reliable high-precision positioning to ensure that the planning and control modules can accurately and quickly complete vehicle maneuvers. Real-time laser positioning is usually used as an auxiliary positioning source in closed campuses to ensure accurate positioning when combined navigation is unreliable or fails. Laser SLAM is widely used in such scene operations, but if there is a lack of loop correction and it is affected by z-axis drift, its global positioning accuracy is usually unreliable and it is difficult to achieve high-precision global positioning. Therefore, real-time laser point cloud matching positioning based on point cloud maps is usually introduced.

[0003] Real-time point cloud matching and positioning mainly include matching and positioning based on static point cloud maps and dynamic and static point cloud maps. Matching and positioning based on static point cloud maps usually periodically updates the map to ensure the reliability and output stability of point cloud matching and positioning. However, in highly dynamic change scenarios, periodic updates are difficult to achieve, and the performance of matching and positioning is greatly reduced in highly dynamic change scenarios, and it is basically impossible to provide effective and reliable high-precision positioning. Summary of the invention

[0004] In view of the above problems, the present invention provides a method, system and storage medium for updating a dynamic point cloud map of a port, which can not only avoid environmental changes and map update delays, reduce their interference with current matching and positioning, and output reliable matching and positioning results in real time, but also minimize the use of controller storage space by the point cloud map.

[0005] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows:

[0006] A method for updating a port dynamic point cloud map, the method comprising:

[0007] U1. The vehicle performs port operation tasks, obtains data information of the original global point cloud map of the port, obtains point cloud data information of the road in real time based on the on-board laser radar, obtains data information of the vehicle's posture in real time based on the on-board IMU, and obtains data information of the vehicle's absolute posture in real time based on the on-board inertial navigation sensor;

[0008] U2. Based on the point cloud data information of the road and the data information of the original global point cloud map of the port, an adaptive upsampling algorithm based on a chaotic mapping sequence is used to extract local features of the point cloud map, and a real-time dense local point cloud map is constructed to obtain data information of the real-time dense local point cloud map;

[0009] U3. Based on the data information of the real-time dense local point cloud map, the data information of the vehicle's posture and the data information of the vehicle's absolute posture, a point cloud key frame matching algorithm based on the vehicle's posture is used to match the key frame of the local point cloud map to obtain data information of the point cloud key frame;

[0010] U4. Based on the data information of the point cloud keyframe, a point cloud map update algorithm based on a graph optimization function is adopted to dynamically update the original global point cloud map of the port to obtain the data information of the updated point cloud map.

[0011] Furthermore, in step U2, the extraction of local features of the point cloud map using an adaptive upsampling algorithm based on a chaotic mapping sequence includes:

[0012] U21. Based on the point cloud data information of the road, a chaotic mapping function Q of the point cloud of the road is established.

[0013]

[0014] Wherein, x is the point cloud data information of the road, α1, α2 and α3 are chaotic mapping factors, and the chaotic mapping sequence of the point cloud of the road is characterized to obtain the data information of the chaotic mapping sequence of the point cloud of the road;

[0015] U22. Based on the data information of the chaotic mapping sequence of the road point cloud and the data information of the original global point cloud map of the port, construct an upsampling function W of the point cloud map,

[0016]

[0017] Among them, y1 is the data information of the chaotic mapping sequence of the road point cloud, y2 is the data information of the original global point cloud map of the port, β1, β2 and β3 are the upsampling factors of the point cloud map;

[0018] U23. Based on the upsampling function W of the point cloud map, the local features of the point cloud map are extracted, and a real-time dense local point cloud map is constructed to obtain data information of the real-time dense local point cloud map.

[0019] Furthermore, the upsampling factors β1, β2 and β3 of the point cloud map are,

[0020]

[0021] Among them, y1 is the data information of the chaotic mapping sequence of the road point cloud, and y2 is the data information of the original global point cloud map of the port.

[0022] Furthermore, the constraint function f of the chaotic mapping factors α1, α2 and α3 is,

[0023]

[0024] The value range of the constraint function f is (0,1).

[0025] Furthermore, in step U3, the matching of the key frames of the local point cloud map using the point cloud key frame matching algorithm based on the vehicle posture includes:

[0026] U31. Based on the data information of the vehicle's posture and the data information of the vehicle's absolute posture, establish a vehicle's posture fusion function R,

[0027]

[0028] Among them, z1 is the data information of the vehicle's posture, z2 is the data information of the vehicle's absolute posture, δ1, δ2 and δ3 are the fusion factors of the vehicle's posture, and the vehicle's posture is fused to obtain the fused vehicle's posture data information;

[0029] U32. Based on the fused data information of the vehicle's posture and the data information of the real-time dense local point cloud map, establish a point cloud key frame matching function P of the vehicle's posture,

[0030] Among them, r1 is the data information of the fused vehicle posture, r2 is the data information of the real-time dense local point cloud map, γ1, γ2 and γ3 are the key frame matching factors of the point cloud;

[0031] U33. Based on the point cloud key frame matching function P of the vehicle posture, match the key frame of the local point cloud map to obtain data information of the point cloud key frame.

[0032] Furthermore, the constraints of the fusion factors δ1, δ2 and δ3 of the vehicle posture are as follows:

[0033] Furthermore, the key frame matching factors γ1, γ2 and γ3 of the point cloud are,

[0034]

[0035] Among them, r1 is the data information of the fused vehicle posture, and r2 is the data information of the real-time dense local point cloud map.

[0036] Furthermore, in step U4, the point cloud map update algorithm based on the graph optimization function is used to dynamically update the original global point cloud map of the port, including:

[0037] U41. Based on the data information of the point cloud key frame, establish a point cloud graph optimization function G,

[0038] Among them, g is the data information of the point cloud key frame, η1, η2 and η3 are graph optimization factors, and the graph optimization is performed on the point cloud key frame to obtain the data information of the point cloud key frame after graph optimization;

[0039] U42. Based on the data information of the point cloud key frame after the image optimization, construct a point cloud dynamic map update function H,

[0040]

[0041] Among them, h is the data information of the point cloud key frame after graph optimization, λ1, λ2 and λ3 are the dynamic update factors of the point cloud map;

[0042] U43. Based on the point cloud dynamic map update function H, the original global point cloud map of the port is dynamically updated to obtain data information of the updated point cloud map.

[0043] In order to achieve the above-mentioned purpose and other related purposes, the present invention also provides a port dynamic point cloud map updating system, including a computer device, which is programmed or configured to execute the steps of any one of the port dynamic point cloud map updating methods.

[0044] In order to achieve the above-mentioned purpose and other related purposes, the present invention also provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the port dynamic point cloud map updating methods.

[0045] The present invention has the following positive effects:

[0046] 1. The present invention extracts local features of point cloud maps by adopting an adaptive upsampling algorithm based on chaotic mapping sequences, constructs a real-time dense local point cloud map, and matches the key frames of the local point cloud map by combining a matching algorithm of point cloud key frames based on vehicle posture to obtain data information of point cloud key frames. This can not only avoid environmental changes and map update delays, reduce their interference with current matching and positioning, output reliable matching and positioning results in real time, but also minimize the use of controller storage space by point cloud maps.

[0047] 2. The present invention dynamically updates the original global point cloud map of the port by adopting a point cloud map update algorithm based on graph optimization functions. This not only makes the updating of point cloud maps in high dynamic scenarios no longer strongly dependent on RTK signals, but also ensures stable and reliable output of matching positioning results by constructing a dense local map. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 Schematic diagram of the method flow of the present invention (I);

[0049] Figure 2 Schematic diagram of the method flow of the present invention (II);

[0050] Figure 3 A schematic diagram of the flow of the adaptive upsampling algorithm based on chaotic mapping sequence of the present invention;

[0051] Figure 4 It is a flow chart of the matching algorithm of the point cloud key frame based on the vehicle posture of the present invention;

[0052] Figure 5 It is a flowchart of the point cloud map updating algorithm based on the graph optimization function of the present invention. DETAILED DESCRIPTION

[0053] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0054] Example 1: Figure 1 or Figure 2 As shown, a method for updating a port dynamic point cloud map comprises:

[0055] U1. The vehicle performs port operation tasks, obtains data information of the original global point cloud map of the port, obtains point cloud data information of the road in real time based on the on-board laser radar, obtains data information of the vehicle's posture in real time based on the on-board IMU, and obtains data information of the vehicle's absolute posture in real time based on the on-board inertial navigation sensor;

[0056] U2. Based on the point cloud data information of the road and the data information of the original global point cloud map of the port, an adaptive upsampling algorithm based on a chaotic mapping sequence is used to extract local features of the point cloud map, and a real-time dense local point cloud map is constructed to obtain data information of the real-time dense local point cloud map;

[0057] U3. Based on the data information of the real-time dense local point cloud map, the data information of the vehicle's posture and the data information of the vehicle's absolute posture, a point cloud key frame matching algorithm based on the vehicle's posture is used to match the key frame of the local point cloud map to obtain data information of the point cloud key frame;

[0058] U4. Based on the data information of the point cloud keyframe, a point cloud map update algorithm based on a graph optimization function is adopted to dynamically update the original global point cloud map of the port to obtain the data information of the updated point cloud map.

[0059] In this embodiment, if Figure 3 As shown, in step U2, the extraction of local features of the point cloud map by using an adaptive upsampling algorithm based on a chaotic mapping sequence includes:

[0060] U21. Based on the point cloud data information of the road, a chaotic mapping function Q of the point cloud of the road is established.

[0061]

[0062] Wherein, x is the point cloud data information of the road, α1, α2 and α3 are chaotic mapping factors, and the chaotic mapping sequence of the point cloud of the road is characterized to obtain the data information of the chaotic mapping sequence of the point cloud of the road;

[0063] U22. Based on the data information of the chaotic mapping sequence of the road point cloud and the data information of the original global point cloud map of the port, construct an upsampling function W of the point cloud map,

[0064]

[0065] Among them, y1 is the data information of the chaotic mapping sequence of the road point cloud, y2 is the data information of the original global point cloud map of the port, β1, β2 and β3 are the upsampling factors of the point cloud map;

[0066] U23. Based on the upsampling function W of the point cloud map, the local features of the point cloud map are extracted, and a real-time dense local point cloud map is constructed to obtain data information of the real-time dense local point cloud map.

[0067] In this embodiment, the upsampling factors β1, β2 and β3 of the point cloud map are,

[0068]

[0069] Among them, y1 is the data information of the chaotic mapping sequence of the road point cloud, and y2 is the data information of the original global point cloud map of the port.

[0070] In this embodiment, the constraint function f of the chaotic mapping factors α1, α2 and α3 is,

[0071] The value range of the constraint function f is (0,1).

[0072] In this embodiment, if Figure 4 As shown, in step U3, the matching algorithm of the point cloud key frame based on the vehicle posture is used to match the key frame of the local point cloud map, including:

[0073] U31. Based on the data information of the vehicle's posture and the data information of the vehicle's absolute posture, establish a vehicle's posture fusion function R,

[0074]

[0075] Among them, z1 is the data information of the vehicle's posture, z2 is the data information of the vehicle's absolute posture, δ1, δ2 and δ3 are the fusion factors of the vehicle's posture, and the vehicle's posture is fused to obtain the fused vehicle's posture data information;

[0076] U32. Based on the fused data information of the vehicle's posture and the data information of the real-time dense local point cloud map, establish a point cloud key frame matching function P of the vehicle's posture,

[0077] Among them, r1 is the data information of the fused vehicle posture, r2 is the data information of the real-time dense local point cloud map, γ1, γ2 and γ3 are the key frame matching factors of the point cloud;

[0078] U33. Based on the point cloud key frame matching function P of the vehicle posture, match the key frame of the local point cloud map to obtain data information of the point cloud key frame.

[0079] In this embodiment, the constraints of the fusion factors δ1, δ2 and δ3 of the vehicle posture are:

[0080]

[0081] In this embodiment, the key frame matching factors γ1, γ2 and γ3 of the point cloud are,

[0082]

[0083] Among them, r1 is the data information of the fused vehicle posture, and r2 is the data information of the real-time dense local point cloud map.

[0084] Example 2: Based on the port dynamic point cloud map updating method of Example 1, the present invention is further illustrated and described below.

[0085] like Figure 1 or Figure 2 As shown, a method for updating a port dynamic point cloud map comprises:

[0086] U1. The vehicle performs port operation tasks, obtains data information of the original global point cloud map of the port, obtains point cloud data information of the road in real time based on the on-board laser radar, obtains data information of the vehicle's posture in real time based on the on-board IMU, and obtains data information of the vehicle's absolute posture in real time based on the on-board inertial navigation sensor;

[0087] U2. Based on the point cloud data information of the road and the data information of the original global point cloud map of the port, an adaptive upsampling algorithm based on a chaotic mapping sequence is used to extract local features of the point cloud map, and a real-time dense local point cloud map is constructed to obtain data information of the real-time dense local point cloud map;

[0088] U3. Based on the data information of the real-time dense local point cloud map, the data information of the vehicle's posture and the data information of the vehicle's absolute posture, a point cloud key frame matching algorithm based on the vehicle's posture is used to match the key frame of the local point cloud map to obtain data information of the point cloud key frame;

[0089] U4. Based on the data information of the point cloud keyframe, a point cloud map update algorithm based on a graph optimization function is adopted to dynamically update the original global point cloud map of the port to obtain the data information of the updated point cloud map.

[0090] In this embodiment. Figure 5 As shown, in step U4, the point cloud map update algorithm based on the graph optimization function is used to dynamically update the original global point cloud map of the port, including:

[0091] U41. Based on the data information of the point cloud key frame, establish a point cloud graph optimization function G,

[0092] Among them, g is the data information of the point cloud key frame, η1, η2 and η3 are graph optimization factors, and the graph optimization is performed on the point cloud key frame to obtain the data information of the point cloud key frame after graph optimization;

[0093] U42. Based on the data information of the point cloud key frame after the image optimization, construct a point cloud dynamic map update function H,

[0094]

[0095] Among them, h is the data information of the point cloud key frame after graph optimization, λ1, λ2 and λ3 are the dynamic update factors of the point cloud map;

[0096] U43. Based on the point cloud dynamic map update function H, the original global point cloud map of the port is dynamically updated to obtain data information of the updated point cloud map.

[0097] In this embodiment, the present invention provides a port dynamic point cloud map updating system, including a computer device, which is programmed or configured to execute the steps of any one of the port dynamic point cloud map updating methods.

[0098] In this embodiment, the present invention provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the port dynamic point cloud map updating methods.

[0099] In this embodiment, any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0100] In summary, the present invention can not only avoid the delay of environmental changes and map updates, reduce their interference with the current matching and positioning, and output reliable matching and positioning results in real time, but also minimize the use of controller storage space by the point cloud map.

[0101] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for updating a port dynamic point cloud map, characterized in that: The method comprises: U1. The vehicle performs port operation tasks, obtains data information of the original global point cloud map of the port, obtains point cloud data information of the road in real time based on the on-board laser radar, obtains data information of the vehicle's posture in real time based on the on-board IMU, and obtains data information of the vehicle's absolute posture in real time based on the on-board inertial navigation sensor; U2. Based on the point cloud data information of the road and the data information of the original global point cloud map of the port, an adaptive upsampling algorithm based on a chaotic mapping sequence is used to extract local features of the point cloud map, and a real-time dense local point cloud map is constructed to obtain data information of the real-time dense local point cloud map; U3. Based on the data information of the real-time dense local point cloud map, the data information of the vehicle's posture and the data information of the vehicle's absolute posture, a point cloud key frame matching algorithm based on the vehicle's posture is used to match the key frame of the local point cloud map to obtain data information of the point cloud key frame; U4. Based on the data information of the point cloud keyframe, a point cloud map update algorithm based on a graph optimization function is adopted to dynamically update the original global point cloud map of the port to obtain the data information of the updated point cloud map.

2. The method for updating the port dynamic point cloud map according to claim 1, characterized in that: In step U2, the extraction of local features of the point cloud map using an adaptive upsampling algorithm based on a chaotic mapping sequence includes: U21. Based on the point cloud data information of the road, a chaotic mapping function Q of the point cloud of the road is established. Wherein, x is the point cloud data information of the road, α1, α2 and α3 are chaotic mapping factors, and the chaotic mapping sequence of the point cloud of the road is characterized to obtain the data information of the chaotic mapping sequence of the point cloud of the road; U22. Based on the data information of the chaotic mapping sequence of the road point cloud and the data information of the original global point cloud map of the port, construct an upsampling function W of the point cloud map, Among them, y1 is the data information of the chaotic mapping sequence of the road point cloud, y2 is the data information of the original global point cloud map of the port, β1, β2 and β3 are the upsampling factors of the point cloud map; U23. Based on the upsampling function W of the point cloud map, the local features of the point cloud map are extracted, and a real-time dense local point cloud map is constructed to obtain data information of the real-time dense local point cloud map.

3. The method for updating the port dynamic point cloud map according to claim 2 is characterized in that: The upsampling factors β1, β2 and β3 of the point cloud map are, Among them, y1 is the data information of the chaotic mapping sequence of the road point cloud, and y2 is the data information of the original global point cloud map of the port.

4. The method for updating the port dynamic point cloud map according to claim 2 is characterized in that: The constraint function f of the chaotic mapping factors α1, α2 and α3 is, The value range of the constraint function f is (0,1).

5. The method for updating the port dynamic point cloud map according to claim 1, characterized in that: In step U3, the matching algorithm of the point cloud key frame based on the vehicle posture is used to match the key frame of the local point cloud map, including: U31. Based on the data information of the vehicle's posture and the data information of the vehicle's absolute posture, establish a vehicle's posture fusion function R, Among them, z1 is the data information of the vehicle's posture, z2 is the data information of the vehicle's absolute posture, δ1, δ2 and δ3 are the fusion factors of the vehicle's posture, and the vehicle's posture is fused to obtain the fused vehicle's posture data information; U32. Based on the fused data information of the vehicle's posture and the data information of the real-time dense local point cloud map, establish a point cloud key frame matching function P of the vehicle's posture, Among them, r1 is the data information of the fused vehicle posture, r2 is the data information of the real-time dense local point cloud map, γ1, γ2 and γ3 are the key frame matching factors of the point cloud; U33. Based on the point cloud key frame matching function P of the vehicle posture, the key frame of the local point cloud map is matched to obtain the data information of the point cloud key frame.

6. The method for updating the port dynamic point cloud map according to claim 5, characterized in that: The constraints of the fusion factors δ1, δ2 and δ3 of the vehicle posture are:

7. The method for updating the port dynamic point cloud map according to claim 5, characterized in that: The key frame matching factors γ1, γ2 and γ3 of the point cloud are, Among them, r1 is the data information of the fused vehicle posture, and r2 is the data information of the real-time dense local point cloud map.

8. The method for updating the port dynamic point cloud map according to claim 1, characterized in that: In step U4, the point cloud map update algorithm based on the graph optimization function is used to dynamically update the original global point cloud map of the port, including: U41. Based on the data information of the point cloud key frame, establish a point cloud graph optimization function G, Among them, g is the data information of the point cloud key frame, η1, η2 and η3 are graph optimization factors, and the graph optimization is performed on the point cloud key frame to obtain the data information of the point cloud key frame after graph optimization; U42. Based on the data information of the point cloud key frame after the image optimization, construct a point cloud dynamic map update function H, Among them, h is the data information of the point cloud key frame after graph optimization, λ1, λ2 and λ3 are the dynamic update factors of the point cloud map; U43. Based on the point cloud dynamic map update function H, the original global point cloud map of the port is dynamically updated to obtain data information of the updated point cloud map.

9. A port dynamic point cloud map updating system, comprising a computer device, characterized in that: The computer device is programmed or configured to execute the steps of the port dynamic point cloud map updating method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program programmed or configured to execute the port dynamic point cloud map updating method described in any one of claims 1 to 8.

Images