WO2022143360A1 - 一种环境地图自主更新方法、设备及计算机可读存储介质 - Google Patents

一种环境地图自主更新方法、设备及计算机可读存储介质 Download PDF

Info

Publication number
WO2022143360A1
WO2022143360A1 PCT/CN2021/140658 CN2021140658W WO2022143360A1 WO 2022143360 A1 WO2022143360 A1 WO 2022143360A1 CN 2021140658 W CN2021140658 W CN 2021140658W WO 2022143360 A1 WO2022143360 A1 WO 2022143360A1
Authority
WO
WIPO (PCT)
Prior art keywords
map
robot
pose
point cloud
environment
Prior art date
Application number
PCT/CN2021/140658
Other languages
English (en)
French (fr)
Inventor
张米令
刘俊斌
穆星元
Original Assignee
炬星科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 炬星科技(深圳)有限公司 filed Critical 炬星科技(深圳)有限公司
Publication of WO2022143360A1 publication Critical patent/WO2022143360A1/zh

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/005Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20Instruments for performing navigational calculations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23Updating
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29Geographical information databases

Definitions

  • the present invention relates to the field of robotics, and in particular, to a method, device and computer-readable storage medium for autonomously updating an environment map.
  • the map previously constructed by the robot cannot fully describe the current operating environment of the robot. At this time, if you continue to use the old environmental map for positioning, it may affect the positioning accuracy and navigation efficiency of the robot. If it is rebuilt, it will greatly increase the maintenance cost of the system, increase human and material resources, and reduce production efficiency. .
  • the present invention provides a method, device and computer-readable storage medium for autonomously updating an environmental map, so as to solve the problem that the robot cannot update the map dynamically and autonomously when constructing a map, resulting in the influence of the old map on positioning and navigation, and in a dynamic environment.
  • the stability and accuracy of the operation are not high.
  • the present invention provides a method for autonomously updating an environment map, the method comprising:
  • the third grid map is deleted from the updated map.
  • the present invention also provides a device for autonomously updating an environment map, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being executed by the processor
  • a device for autonomously updating an environment map which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being executed by the processor
  • the present invention also provides a computer-readable storage medium, on which an autonomous environment map update program is stored, and when the environment map autonomous update program is executed by the processor, the autonomous environment map as described in any of the above is realized. Steps to update the method.
  • the environment map can be updated independently, which improves the stability and accuracy of the robot's operation in a dynamic environment, avoids the influence of old maps on positioning and navigation, and avoids the extra cost caused by re-building the map, which improves production. efficiency.
  • Fig. 1 is the flow chart of the first embodiment of the method for autonomous updating of the environment map of the present invention
  • Fig. 2 is a flow chart of the second embodiment of the method for autonomously updating an environment map of the present invention
  • FIG. 3 is a flowchart of a third embodiment of the method for autonomously updating an environment map of the present invention.
  • FIG. 4 is a flowchart of a fourth embodiment of the method for autonomously updating an environment map of the present invention.
  • FIG. 5 is a flowchart of the fifth embodiment of the method for autonomously updating an environment map of the present invention.
  • FIG. 6 is a flowchart of a sixth embodiment of the method for autonomously updating an environment map of the present invention.
  • FIG. 7 is a flowchart of a seventh embodiment of the method for autonomously updating an environment map of the present invention.
  • FIG. 8 is a flowchart of an eighth embodiment of the method for autonomously updating an environment map of the present invention.
  • FIG. 9 is a schematic diagram of a flow chart of an autonomous update of an environment map according to the first embodiment of the method for autonomous update of an environment map of the present invention.
  • FIG. 10 is a schematic diagram of the autonomous updating and positioning of the environmental map according to the first embodiment of the method for autonomously updating the environmental map of the present invention.
  • FIG. 11 is a schematic diagram of the pose of the environment map autonomously updating according to the seventh embodiment of the environment map autonomous updating method of the present invention.
  • FIG. 1 is a flow chart of the first embodiment of the method for autonomously updating an environment map according to the present invention.
  • a method for autonomously updating an environment map comprising:
  • a map update scheme based on an existing map is proposed. It can run normally and avoid problems that affect the positioning and navigation of the robot;
  • the map is updated for the surrounding environment, During the period, constantly perceive the latest surrounding environment to obtain the latest information of the current environment.
  • the information in the old map will be replaced.
  • the latest map will be saved, and then the robot will use The latest map for positioning and navigation;
  • a factory application engineer can construct a map of the environment around which the robot operates.
  • this embodiment adopts simultaneous positioning based on map optimization. with map building techniques.
  • the constructed map saves the pose of the robot and the sensor data of the robot under the pose.
  • the sensor data includes lidar, inertial measurement unit (imu) and encoder.
  • the mapping link the classic graph optimization synchronous positioning and mapping (slam) framework is used, and the iterative closest point method (ICP, Iterative method) is used in the front end. Closest Points Algorithm algorithm for data association to build a pose graph graph), and finally, loop closure detection is performed at the back end, and the pose of the robot is optimized by nonlinear least squares.
  • the current pose map (pose graph) and the saved historical pose graph (pose graph) to get the latest position of the current robot.
  • the beneficial effect of this embodiment is that, by recognizing that when the robot runs to a historical position, the point cloud at the current moment and the point cloud at the historical moment are spliced to construct a loopback constraint to obtain a pose graph; the pose graph is optimized. , obtain the optimal robot pose; then, collect sensor data under the optimal robot pose to generate an updated map; finally, in the updated map, the sensor data includes The point cloud of the robot is spliced with the point cloud of the historical moment to obtain the current pose of the robot.
  • a solution capable of autonomously updating the environment map is realized, which improves the stability and accuracy of the robot's operation in a dynamic environment, avoids the influence of the old map on positioning and navigation, and avoids the problems caused by re-mapping. Additional costs and increased production efficiency.
  • Fig. 2 is the flow chart of the second embodiment of the environmental map autonomous updating method of the present invention. Based on the above-mentioned embodiment, when the robot runs to a historical position, the point cloud at the current moment and the point cloud at the historical moment are spliced, and a loopback constraint is constructed, Before getting the pose graph, include:
  • the process of map building is performed in the operating environment
  • the historical pose of the robot and the sensor data under the historical pose are saved;
  • a map of the operating environment is constructed by using the sensor data.
  • the beneficial effect of this embodiment is that, by determining the operating environment of the robot; then, in the process of executing map construction in the operating environment, the historical pose of the robot and the sensor data under the historical pose are saved .
  • a solution capable of autonomously updating the environment map is realized, which provides the map information base of the operating environment, improves the stability and accuracy of the robot's operation in a dynamic environment, and avoids the influence of old maps on positioning and navigation. Avoid the extra cost of re-mapping and improve production efficiency.
  • FIG. 3 is a flowchart of a third embodiment of the method for autonomously updating an environment map according to the present invention. Based on the above-mentioned embodiment, when the robot runs to a historical position, the point cloud at the current moment and the point cloud at the historical moment are spliced to construct a loopback constraint, Before getting the pose graph, it also includes:
  • the operating environment has completed the map construction, when any robot enters the operating environment, it can be determined to perform map optimization in the operating environment while positioning and map construction;
  • the same robot when the same robot enters the operating environment multiple times, it can perform map optimization and map construction multiple times in the operating environment, so as to maintain the map in the environment and the environment.
  • the dynamic update is consistent.
  • the beneficial effect of this embodiment is that, by identifying whether the operation environment of the robot has completed the map construction; if the operation environment has completed the map construction, it is determined to perform map optimization in the operation environment while positioning and mapping Construct.
  • a solution capable of autonomously updating the environment map is realized, providing a way to update the map information of the operating environment, improving the stability and accuracy of the robot's operation in a dynamic environment, and avoiding the influence of old maps on positioning and navigation. It also avoids the extra cost of re-mapping and improves production efficiency.
  • FIG. 4 is a flowchart of a fourth embodiment of the method for autonomously updating an environment map according to the present invention. Based on the above embodiment, when the robot runs to a historical position, the point cloud at the current moment and the point cloud at the historical moment are spliced to construct a loopback constraint, to get the pose graph, including:
  • an iterative closest point method (ICP, Iterative The Closest Points Algorithm) algorithm splices the laser point clouds collected by the robot's laser sensor at different times to obtain the pose changes of the robot at two times;
  • the above-mentioned pose includes a rotation matrix R and a translation vector t;
  • a successful laser data splicing constructs a pose constraint of the robot at different times, that is, it can be regarded as constructing an edge of the pose graph;
  • the poses of the robot at different times constitute each node in the pose graph
  • the robot in the process of building a map, continuously keeps up with the laser data of a moment, and performs splicing, so as to construct a constraint edge similar to an odometer.
  • the beneficial effect of this embodiment is that, by recognizing that when the robot runs in the operating environment again, new sensor data is continuously acquired; The sensor data is stitched to build constrained edges.
  • a solution capable of autonomously updating the environment map is realized, which provides the basis for determining the constraint edge, improves the stability and accuracy of the robot's operation in a dynamic environment, avoids the influence of old maps on positioning and navigation, and avoids The extra cost caused by re-mapping is eliminated, and the production efficiency is improved.
  • FIG. 5 is a flowchart of a fifth embodiment of the method for autonomously updating an environment map according to the present invention. Based on the above embodiment, when the robot runs to a historical position, the point cloud at the current moment and the point cloud at the historical moment are spliced to construct a loopback constraint, to get the pose graph, which also includes:
  • the robot when the robot reaches a previously reached position, it can be spliced with the point cloud at an earlier time to construct a loopback constraint;
  • a pose graph including odometer edges and loopback edges is obtained according to the above constraints.
  • the beneficial effect of this embodiment is that, by recognizing that when the robot runs to the historical position of the historical map, the point cloud at the current moment and the point cloud at the historical moment are spliced to construct the loop closure constraint; Constraining edges and the pose graph of the loop closure constraints. It realizes a scheme that can update the environment map autonomously, provides the generation scheme of the pose graph, improves the stability and accuracy of the robot in a dynamic environment, avoids the influence of the old map on positioning and navigation, and also Avoid the extra cost of re-mapping and improve production efficiency.
  • FIG. 6 is a flowchart of the sixth embodiment of the method for autonomously updating an environment map according to the present invention. Based on the above embodiment, in the updated map, the point cloud included in the sensor data and the historical time points are updated. After the cloud is spliced to obtain the current pose of the robot, it includes:
  • S52 Acquire maps of the current multiple pose nodes of the robot.
  • a least squares optimization algorithm is used to optimize the above pose graph to obtain the optimal robot pose
  • a 2D grid map of the operating environment is obtained by using the optimal robot pose and the robot laser sensor data collected under the pose;
  • a preset overlap threshold for determining the update of the old and new maps is preset, and then the maps of multiple pose nodes are performed according to the overlap threshold. overlap determination.
  • the beneficial effect of this embodiment is that the overlapping threshold used to determine the update of the old and new maps is preset; then, the maps of the current multiple pose nodes of the robot are acquired.
  • a solution capable of autonomously updating the environment map is implemented, providing an overlap threshold for judging the update of the old and new maps, improving the stability and accuracy of the robot's operation in a dynamic environment, and avoiding the old maps for positioning and navigation. It also avoids the extra cost caused by re-mapping and improves production efficiency.
  • FIG. 7 is a flowchart of a seventh embodiment of the method for autonomously updating an environment map according to the present invention. Based on the above embodiment, in the updated map, the point cloud included in the sensor data and the historical time points are updated. After the cloud is spliced to obtain the current pose of the robot, it also includes:
  • the schematic diagram of the pose of the environment map is automatically updated. Specifically, in the map update process, in order to reduce the size of the updated map, it is necessary to remove the old map. .
  • the old laser nodes are eliminated according to the size of the overlapping area between the robot poses.
  • the first grid map rendered by the current first pose node node1 and the second grid rendered by the second pose node node2 of the robot are sequentially acquired.
  • the first non-overlapping area of the first grid map and the third grid map is acquired, and the difference between the second grid map and the third grid map is acquired.
  • obtain the area of the overlapping area of the first non-overlapping area and the second non-overlapping area that is, obtaining the first grid map and the second grid map together with the first grid map.
  • the area of the non-overlapping area of the three-grid map that is, the area of the black block-shaped area shown in Figure 11, is used to determine the degree of overlap of the maps of multiple pose nodes. The higher the overlap degree of , the smaller the area is, and conversely, the lower the overlap degree of the maps of multiple pose nodes, the larger the area.
  • the beneficial effect of this embodiment is that the first grid map rendered by the current first pose node, the second grid map rendered by the second pose node, and the third pose node rendering of the robot are sequentially acquired. the third grid map; then, obtain the area of the non-overlapping area of the first grid map, the second grid map and the third grid map.
  • a solution capable of autonomously updating environmental maps is realized, providing a way of judging the degree of overlap, improving the stability and accuracy of the robot's operation in a dynamic environment, avoiding the influence of old maps on positioning and navigation, and avoiding The additional cost of re-mapping has been eliminated, and the production efficiency has been improved.
  • FIG. 8 is a flowchart of an eighth embodiment of the method for autonomously updating an environment map according to the present invention. Based on the above embodiment, in the updated map, the point cloud included in the sensor data and the historical time points are updated. After the cloud is spliced to obtain the current pose of the robot, it also includes:
  • the area is smaller than the overlap threshold, it is determined that the degree of overlap is high, and at this time, the newly acquired third grid map is deleted from the updated map. , thus keeping the memory footprint of the map relatively stable;
  • the corresponding overlap threshold is determined according to different dynamic environments, and when the variation of the dynamic environment is relatively high, in order to improve the map update efficiency, a higher overlap threshold may be set;
  • the corresponding overlapping thresholds are determined according to different navigation or positioning accuracy requirements, and when the navigation or positioning accuracy requirements are high, a lower overlapping threshold may be set;
  • corresponding overlap thresholds are determined according to different memory occupancy limits, and when the memory occupancy limit is higher, a lower overlap threshold may be set.
  • the beneficial effect of this embodiment is that by judging whether the area is smaller than the overlap threshold; then, if the area is smaller than the overlap threshold, the third grid map is deleted from the updated map .
  • the present invention also proposes a device for autonomously updating an environment map.
  • the device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being When executed, the processor implements the steps of the method for autonomously updating an environment map as described in any one of the above.
  • the present invention also proposes a computer-readable storage medium, on which an environmental map autonomous update program is stored, and when the environmental map autonomous update program is executed by the processor, any of the above-mentioned steps are implemented.
  • a computer-readable storage medium on which an environmental map autonomous update program is stored, and when the environmental map autonomous update program is executed by the processor, any of the above-mentioned steps are implemented. The steps of the self-updating method for the environment map described above.
  • the method, device and computer-readable storage medium for autonomously updating an environment map when the robot runs to a historical position, the point cloud at the current moment and the point cloud at the historical moment are spliced together, and a loopback constraint is constructed to obtain a pose graph; Optimizing the pose graph to obtain the optimal robot pose; then, collecting sensor data under the optimal robot pose to generate an updated map; finally, in the updated map , splicing the point cloud contained in the sensor data with the point cloud at the historical moment to obtain the current pose of the robot.
  • the environment map can be updated autonomously, the stability and accuracy of the robot's operation in a dynamic environment can be improved, and the influence of the old map on positioning and navigation can be avoided.
  • it can also avoid the extra cost caused by re-creating the map and improve the production efficiency. . Therefore, it has industrial applicability.

Landscapes

  • Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

一种环境地图自主更新方法、设备及计算机可读存储介质,其中,方法包括:当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图(S1);对位姿图进行优化,得到最优的机器人位姿(S2);在最优的机器人位姿下采集传感器数据,以生成更新后的地图(S3);在更新后的地图中,对传感器数据中包含的点云与历史时刻点云进行拼接,以得到机器人的当前位姿(S4)。实现了一种能够进行环境地图自主更新的方案,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。

Description

一种环境地图自主更新方法、设备及计算机可读存储介质
本发明涉及机器人技术领域,尤其涉及一种环境地图自主更新方法、设备及计算机可读存储介质。
背景技术
现有技术中,针对移动机器人的实际应用,通常需要对机器人运行的环境进行地图构建。由此,机器人在已构建地图的场景中运行时,会基于构建的环境地图进行定位以及导航。理论上,大多数机器人应用的过程中,常假设机器人运行的环境是静态的,不变的。然而,在实际机器人的部署应用过程中,其所处的环境常常是动态改变的。
如上所述,若机器人运行的环境发生了变化,那么机器人先前构建的地图并不能完全描述机器人当前所处的运行环境。此时,如果继续利用旧的环境地图进行定位,可能会影响到机器人定位精度以及导航效率,而若重新建,则又会较大的增加系统的维护成本,增加了人力物力,降低了生产效率。
技术问题
有鉴于此,本发明提供一种一种环境地图自主更新方法、设备及计算机可读存储介质,以解决机器人建图时不能动态自主更新地图,导致旧地图对定位和导航的影响而动态环境下运行的稳定性和准确性不高的问题。
技术解决方案
本发明提出了一种环境地图自主更新方法,该方法包括:
当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图;
对所述位姿图进行优化,得到最优的机器人位姿;
在所述最优的机器人位姿下采集传感器数据,以生成更新后的地图;
在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿;
预设用于判定新旧地图更新的重叠阈值;
获取所述机器人的当前的多个位姿节点的地图;
依次获取所述机器人的当前的第一位姿节点渲染的第一网格地图、第二位姿节点渲染的第二网格地图、以及第三位姿节点渲染的第三网格地图;
获取所述第一网格地图与所述第三网格地图的第一非重合区域,获取所述第二网格地图与所述第三网格地图的第二非重合区域,获取所述第一非重合区域与所述第二非重合区域的重合区域的面积;判断所述面积是否小于所述重叠阈值;
若所述面积小于所述重叠阈值,则将所述第三网格地图从所述更新后的地图中删除。
本发明还提出了一种环境地图自主更新设备,该设备包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现如上任一项所述的环境地图自主更新方法的步骤。
本发明还提出了一种计算机可读存储介质,该计算机可读存储介质上存储有环境地图自主更新程序,环境地图自主更新程序被处理器执行时实现如上述任一项所述的环境地图自主更新方法的步骤。
有益效果
实施本发明的环境地图自主更新方法、设备及计算机可读存储介质,当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图;对所述位姿图进行优化,得到最优的机器人位姿;然后,在所述最优的机器人位姿下采集传感器数据,以生成更新后的地图;最后,在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿。从而实现环境地图自主更新,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。
附图说明
下面将结合附图及实施例对本发明作进一步说明,附图中:
图1是本发明环境地图自主更新方法第一实施例的流程图;
图2是本发明环境地图自主更新方法第二实施例的流程图;
图3是本发明环境地图自主更新方法第三实施例的流程图;
图4是本发明环境地图自主更新方法第四实施例的流程图;
图5是本发明环境地图自主更新方法第五实施例的流程图;
图6是本发明环境地图自主更新方法第六实施例的流程图;
图7是本发明环境地图自主更新方法第七实施例的流程图;
图8是本发明环境地图自主更新方法第八实施例的流程图;
图9是本发明环境地图自主更新方法第一实施例的环境地图自主更新流程示意图;
图10是本发明环境地图自主更新方法第一实施例的环境地图自主更新定位示意图;
图11是本发明环境地图自主更新方法第七实施例的环境地图自主更新的位姿示意图。
本发明的实施方式
应当理解,此处所描述的具体实施例仅用以解释本发明,并不用于限定本发明。
在后续的描述中,使用用于表示元件的诸如“模块”、“部件”或“单元”的后缀仅为了有利于本发明的说明,其本身没有特定的意义。因此,“模块”、“部件”或“单元”可以混合地使用。
实施例一
图1是本发明环境地图自主更新方法第一实施例的流程图。一种环境地图自主更新方法,该方法包括:
S1、当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图;
S2、对所述位姿图进行优化,得到最优的机器人位姿;
S3、在所述最优的机器人位姿下采集传感器数据,以生成更新后的地图;
S4、在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿。
可选地,为了有效应对机器人运行场景发生变化的情况,在本实施例中,提出了一种基于已有地图的地图更新方案,通过对环境变化进行自主更新,以确保机器人在动态环境中仍然可以正常运行,避免出现影响机器人定位以及导航的问题;
可选地,在本实施例中,参考图9示出的环境地图自主更新流程示意图,在机器人运行前,已存在构建的地图,在机器人再次运行过程中,首先,对周围环境进行地图更新,期间,不断地感知周围最新的环境,以获取当前环境的最新信息,可选地,为了减小资源占用,将替换掉老旧地图中的信息,最后,将最新的地图保存,之后机器人将采用最新的地图进行定位导航;
可选地,在本实施例中,具体的,机器人在部署过程中,首先,可由工厂应用工程师对机器人运行周围环境进行地图构建,在地图构建过程中,本实施例采用基于图优化的同时定位与地图构建技术。其中,构建的地图中保存了机器人的位姿以及该位姿下机器人的传感器数据。其中,传感器数据包含激光雷达,惯性测量单元(imu)以及编码器。然后,在建图环节采用经典的图优化同步定位与建图(slam)框架,在前端利用迭代最近点法(ICP,Iterative Closest Points Algorithm)算法进行数据关联,以构建位姿图(pose graph),最后,在后端进行回环检测,非线性最小二乘优化机器人的位姿。
可选地,在本实施例中,参考图10示出的环境地图自主更新定位示意图,在定位阶段,将当前的位姿图(pose graph)与已经保存的历史的位姿图(pose graph进行匹配,以获取当前机器人最新位置。
本实施例的有益效果在于,通过识别到当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图;对所述位姿图进行优化,得到最优的机器人位姿;然后,在所述最优的机器人位姿下采集传感器数据,以生成更新后的地图;最后,在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿。实现了一种能够进行环境地图自主更新的方案,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。
实施例二
图2是本发明环境地图自主更新方法第二实施例的流程图,基于上述实施例,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图之前,包括:
S01、确定所述机器人的运行环境;
S02、在所述运行环境下执行地图构建的过程中,保存所述机器人的历史位姿以及所述历史位姿下的传感器数据。
可选地,在本实施例中,当所述机器人的运行环境为未建图环境时,在所述运行环境下执行地图构建的过程;
可选地,在本实施例中,在所述运行环境下执行地图构建的过程中,保存所述机器人的历史位姿以及所述历史位姿下的传感器数据;
可选地,在本实施例中,通过所述传感器数据构建运行环境的地图。
本实施例的有益效果在于,通过确定所述机器人的运行环境;然后,在所述运行环境下执行地图构建的过程中,保存所述机器人的历史位姿以及所述历史位姿下的传感器数据。实现了一种能够进行环境地图自主更新的方案,提供了运行环境的地图信息基础,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。
实施例三
图3是本发明环境地图自主更新方法第三实施例的流程图,基于上述实施例,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图之前,还包括:
S03、识别所述机器人的所述运行环境是否已完成地图构建;
S04、若所述运行环境已完成地图构建,则确定在所述运行环境下执行图优化的同时定位与地图构建。
可选地,在本实施例中,若所述运行环境已完成地图构建,则在任一机器人进入该运行环境时,即可确定在所述运行环境下执行图优化的同时定位与地图构建;
可选地,在本实施例中,多个机器人进入该运行环境时,可同时在所述运行环境下执行图优化的同时定位与地图构建;
可选地,在本实施例中,同一机器人多次进入该运行环境时,即可多次在所述运行环境下执行图优化的同时定位与地图构建,从而保持该环境下的地图与该环境的动态更新保持一致。
本实施例的有益效果在于,通过识别所述机器人的所述运行环境是否已完成地图构建;若所述运行环境已完成地图构建,则确定在所述运行环境下执行图优化的同时定位与地图构建。实现了一种能够进行环境地图自主更新的方案,提供了运行环境的地图信息更新途径,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。 
实施例四
图4是本发明环境地图自主更新方法第四实施例的流程图,基于上述实施例,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图,包括:
S11、当所述机器人再次运行于所述运行环境时,持续地获取新的所述传感器数据;
S12、连续地对前、后时刻获取到的新的所述传感器数据进行拼接,以构建约束边。
可选地,在本实施例中,利用迭代最近点法(ICP,Iterative Closest Points Algorithm)算法对不同时刻机器人激光传感器采集到的激光点云进行拼接,得到两时刻机器人的位姿变化;
可选地,在本实施例中,上述位姿包括旋转矩阵R和已经平移向量t;
可选地,在本实施例中,一次成功的激光数据拼接,构建了机器人不同时刻间的一个位姿约束,也即,可看作是构建了位姿图的一条边;
可选地,在本实施例中,针对不同时刻机器人的位姿构成了位姿图中的各个节点;
可选地,在本实施例中,机器人在建图过程中,不断地跟上一个时刻的激光数据,进行拼接,用以构建类似于里程计式的约束边。
本实施例的有益效果在于,通过识别到当所述机器人再次运行于所述运行环境时,持续地获取新的所述传感器数据;然后,连续地对前、后时刻获取到的新的所述传感器数据进行拼接,以构建约束边。实现了一种能够进行环境地图自主更新的方案,提供了约束边的确定基础,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。  
实施例五
图5是本发明环境地图自主更新方法第五实施例的流程图,基于上述实施例,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图,还包括:
S13、当所述机器人运行至历史地图的历史位置时,对当前时刻点云和历史时刻点云进行拼接,以构建所述回环约束;
S14、生成包含所述约束边和所述回环约束的所述位姿图。
可选地,在本实施例中,当机器人到达先前已经到达的位置时,可以跟较早时刻的点云进行拼接,构建回环约束;
可选地,在本实施例中,根据上述约束得到包含里程计边以及回环边的位姿图。
本实施例的有益效果在于,通过识别到当所述机器人运行至历史地图的历史位置时,对当前时刻点云和历史时刻点云进行拼接,以构建所述回环约束;然后,生成包含所述约束边和所述回环约束的所述位姿图。实现了一种能够进行环境地图自主更新的方案,提供了位姿图的生成方案,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。
实施例六
图6是本发明环境地图自主更新方法第六实施例的流程图,基于上述实施例,所述在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿之后,包括:
S51、预设用于判定新旧地图更新的重叠阈值;
S52、获取所述机器人的当前的多个位姿节点的地图。
可选地,在本实施例中,在此步骤之前,利用最小二乘优化算法,对上述位姿图进行优化,得到最优的机器人位姿;
可选地,在本实施例中,得到最优的机器人位姿之后,利用最优的机器人位姿以及该位姿下采集到的机器人激光传感器数据,得到运行环境的2D的栅格地图;
可选地,在本实施例中,为了避免重复的老旧地图占用过多的内存,将预设用于判定新旧地图更新的重叠阈值,然后,通过该重叠阈值进行多个位姿节点的地图的重叠度判定。
本实施例的有益效果在于,通过预设用于判定新旧地图更新的重叠阈值;然后,获取所述机器人的当前的多个位姿节点的地图。实现了一种能够进行环境地图自主更新的方案,提供了用于判定新旧地图更新的重叠阈值,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。
实施例七
图7是本发明环境地图自主更新方法第七实施例的流程图,基于上述实施例,所述在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿之后,还包括:
S53、依次获取所述机器人的当前的第一位姿节点渲染的第一网格地图、第二位姿节点渲染的第二网格地图、以及第三位姿节点渲染的第三网格地图;
S54、获取所述第一网格地图与所述第三网格地图的第一非重合区域,获取所述第二网格地图与所述第三网格地图的第二非重合区域,获取所述第一非重合区域与所述第二非重合区域的重合区域的面积,即获取所述第一网格地图、所述第二网格地图、以及所述第三网格地图的共同重合区域的面积。
可选地,在本实施例中,如图11示出的环境地图自主更新的位姿示意图,具体的,在地图更新过程中,为了减小更新后地图的大小,需要对旧的地图进行剔除。
可选地,在本实施例中,根据机器人位姿之间的重叠区域大小来对旧的激光节点进行剔除。
可选地,在本实施例中,如图11所示,依次获取所述机器人的当前的第一位姿节点node1渲染的第一网格地图、第二位姿节点node2渲染的第二网格地图、以及第三位姿节点node3渲染的第三网格地图。
可选地,在本实施例中,获取所述第一网格地图与所述第三网格地图的第一非重合区域,获取所述第二网格地图与所述第三网格地图的第二非重合区域,获取所述第一非重合区域与所述第二非重合区域的重合区域的面积,即获取所述第一网格地图、所述第二网格地图共同与所述第三网格地图的非重合区域的面积,也即,图11示出的黑色块状区域的面积,以根据该面积判断多个位姿节点的地图的重叠度,若多个位姿节点的地图的重叠度越高,则该面积越小,反之,若多个位姿节点的地图的重叠度越低,则该面积越大。
本实施例的有益效果在于,通过依次获取所述机器人的当前的第一位姿节点渲染的第一网格地图、第二位姿节点渲染的第二网格地图、以及第三位姿节点渲染的第三网格地图;然后,获取所述第一网格地图、所述第二网格地图共同与所述第三网格地图的非重合区域的面积。实现了一种能够进行环境地图自主更新的方案,提供了重叠度的判断方式,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。
实施例八
图8是本发明环境地图自主更新方法第八实施例的流程图,基于上述实施例,所述在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿之后,还包括:
S55、判断所述面积是否小于所述重叠阈值;
S56、若所述面积小于所述重叠阈值,则将所述第三网格地图从所述更新后的地图中删除。
可选地,在本实施例中,若所述面积小于所述重叠阈值,则确定重叠度较高,此时,将新获取的所述第三网格地图从所述更新后的地图中删除,由此保持地图的内存占用相对稳定;
可选地,在本实施例中,根据不同的动态环境确定相应的重叠阈值,当动态环境的变化量较高时,为了提高地图更新效率,可以设定较高的重叠阈值;
可选地,在本实施例中,根据不同的导航或定位的精准度需求确定相应的重叠阈值,当导航或定位的精准度需求较高时,可以设定较低的重叠阈值;
可选地,在本实施例中,根据不同的内存占用限值确定相应的重叠阈值,当内存占用限值较高时,可以设定较低的重叠阈值。
本实施例的有益效果在于,通过判断所述面积是否小于所述重叠阈值;然后,若所述面积小于所述重叠阈值,则将所述第三网格地图从所述更新后的地图中删除。实现了一种能够进行环境地图自主更新的方案,提供了不同的重叠度选用方案,提高了机器人在动态环境下运行的稳定性和准确性,避免了老旧地图对定位和导航的影响,也避免了重新建图所带来的额外成本,提高了生产效率。
实施例九
基于上述实施例,本发明还提出了一种环境地图自主更新设备,该设备包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现如上任一项所述的环境地图自主更新方法的步骤。
需要说明的是,上述设备实施例与方法实施例属于同一构思,其具体实现过程详细见方法实施例,且方法实施例中的技术特征在设备实施例中均对应适用,这里不再赘述。
实施例十
基于上述实施例,本发明还提出了一种计算机可读存储介质,该计算机可读存储介质上存储有环境地图自主更新程序,环境地图自主更新程序被处理器执行时实现如上述任一项所述的环境地图自主更新方法的步骤。
需要说明的是,上述介质实施例与方法实施例属于同一构思,其具体实现过程详细见方法实施例,且方法实施例中的技术特征在介质实施例中均对应适用,这里不再赘述。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
上述本发明实施例序号仅仅为了描述,不代表实施例的优劣。
上面结合附图对本发明的实施例进行了描述,但是本发明并不局限于上述的具体实施方式,上述的具体实施方式仅仅是示意性的,而不是限制性的,本领域的普通技术人员在本发明的启示下,在不脱离本发明宗旨和权利要求所保护的范围情况下,还可做出很多形式,这些均属于本发明的保护之内。
工业实用性
本发明实施例的环境地图自主更新方法、设备及计算机可读存储介质,当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图;对所述位姿图进行优化,得到最优的机器人位姿;然后,在所述最优的机器人位姿下采集传感器数据,以生成更新后的地图;最后,在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿。从而实现环境地图自主更新,提高机器人在动态环境下运行的稳定性和准确性,避免老旧地图对定位和导航的影响,此外,还能避免重新建图所带来的额外成本,提高生产效率。因此,具有工业实用性。

Claims (7)

  1. 一种环境地图自主更新方法,所述方法包括:
    当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图;
    对所述位姿图进行优化,得到最优的机器人位姿;
    在所述最优的机器人位姿下采集传感器数据,以生成更新后的地图;
    在所述更新后的地图中,对所述传感器数据中包含的点云与所述历史时刻点云进行拼接,以得到所述机器人的当前位姿;
    预设用于判定新旧地图更新的重叠阈值;
    获取所述机器人的当前的多个位姿节点的地图;
    依次获取所述机器人的当前的第一位姿节点渲染的第一网格地图、第二位姿节点渲染的第二网格地图、以及第三位姿节点渲染的第三网格地图;
    获取所述第一网格地图与所述第三网格地图的第一非重合区域,获取所述第二网格地图与所述第三网格地图的第二非重合区域,获取所述第一非重合区域与所述第二非重合区域的重合区域的面积;
    判断所述面积是否小于所述重叠阈值;
    若所述面积小于所述重叠阈值,则将所述第三网格地图从所述更新后的地图中删除。
  2. 根据权利要求1所述的环境地图自主更新方法,其中,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图之前,包括:
    确定所述机器人的运行环境;
    在所述运行环境下执行地图构建的过程中,保存所述机器人的历史位姿以及所述历史位姿下的传感器数据。
  3. 根据权利要求2所述的环境地图自主更新方法,其中,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图之前,还包括:
    识别所述机器人的所述运行环境是否已完成地图构建;
    若所述运行环境已完成地图构建,则确定在所述运行环境下执行图优化的同时定位与地图构建。
  4. 根据权利要求3所述的环境地图自主更新方法,其中,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图,包括:
    当所述机器人再次运行于所述运行环境时,持续地获取新的所述传感器数据;
    连续地对前、后时刻获取到的新的所述传感器数据进行拼接,以构建约束边。
  5. 根据权利要求4所述的环境地图自主更新方法,其中,所述当机器人运行至历史位置时,对当前时刻点云和历史时刻点云进行拼接,构建回环约束,以得到位姿图,还包括:
    当所述机器人运行至历史地图的历史位置时,对当前时刻点云和历史时刻点云进行拼接,以构建所述回环约束;
    生成包含所述约束边和所述回环约束的所述位姿图。
  6. 一种环境地图自主更新设备,所述设备包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现如权利要求1至5中任一项所述的环境地图自主更新方法的步骤。
  7. 一种计算机可读存储介质,,所述计算机可读存储介质上存储有环境地图自主更新程序,所述环境地图自主更新程序被处理器执行时实现如权利要求1至5中任一项所述的环境地图自主更新方法的步骤。
PCT/CN2021/140658 2021-01-04 2021-12-23 一种环境地图自主更新方法、设备及计算机可读存储介质 WO2022143360A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110001149.1 2021-01-04
CN202110001149.1A CN112325873B (zh) 2021-01-04 2021-01-04 一种环境地图自主更新方法、设备及计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2022143360A1 true WO2022143360A1 (zh) 2022-07-07

Family

ID=74302093

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/140658 WO2022143360A1 (zh) 2021-01-04 2021-12-23 一种环境地图自主更新方法、设备及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN112325873B (zh)
WO (1) WO2022143360A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115200572A (zh) * 2022-09-19 2022-10-18 季华实验室 三维点云地图构建方法、装置、电子设备及存储介质
CN115685223A (zh) * 2022-12-15 2023-02-03 深圳市智绘科技有限公司 位置识别方法、装置、电子设备及可读存储介质
CN117576200A (zh) * 2024-01-15 2024-02-20 山东大学 一种长周期移动机器人定位方法、系统、设备及介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112325873B (zh) * 2021-01-04 2021-04-06 炬星科技(深圳)有限公司 一种环境地图自主更新方法、设备及计算机可读存储介质
CN116019562A (zh) * 2021-03-09 2023-04-28 武汉联影智融医疗科技有限公司 机器人控制系统和方法
CN114445359B (zh) * 2022-01-24 2023-09-08 上海歆广数据科技有限公司 一种网格数据体系中电子地图质量控制系统及方法
CN114474065A (zh) * 2022-03-04 2022-05-13 美智纵横科技有限责任公司 机器人控制方法及其装置、机器人和存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107796397A (zh) * 2017-09-14 2018-03-13 杭州迦智科技有限公司 一种机器人双目视觉定位方法、装置和存储介质
CN109709801A (zh) * 2018-12-11 2019-05-03 智灵飞(北京)科技有限公司 一种基于激光雷达的室内无人机定位系统及方法
KR101988555B1 (ko) * 2017-12-05 2019-06-12 충북대학교 산학협력단 조명 불변 영상을 사용한 slam 시스템 및 점구름 지도 생성 방법
CN110689622A (zh) * 2019-07-05 2020-01-14 电子科技大学 一种基于点云分割匹配闭环校正的同步定位与构图算法
CN111553844A (zh) * 2020-04-29 2020-08-18 北京百度网讯科技有限公司 用于更新点云的方法及装置
CN112068154A (zh) * 2020-09-14 2020-12-11 中科院软件研究所南京软件技术研究院 一种激光建图定位方法、装置、存储介质及电子设备
CN112325873A (zh) * 2021-01-04 2021-02-05 炬星科技(深圳)有限公司 一种环境地图自主更新方法、设备及计算机可读存储介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110575653A (zh) * 2018-06-08 2019-12-17 北京眸视科技有限公司 一种智能捡网球机器人
CN108898630B (zh) * 2018-06-27 2020-12-15 清华-伯克利深圳学院筹备办公室 一种三维重建方法、装置、设备和存储介质
CN109074085B (zh) * 2018-07-26 2021-11-09 达闼机器人有限公司 一种自主定位和地图建立方法、装置和机器人
CN111444852A (zh) * 2020-03-27 2020-07-24 深圳市银星智能科技股份有限公司 一种回环检测方法、装置及机器人
CN112014857B (zh) * 2020-08-31 2023-04-07 上海宇航系统工程研究所 用于智能巡检的三维激光雷达定位导航方法及巡检机器人
CN112102400B (zh) * 2020-09-15 2022-08-02 上海云绅智能科技有限公司 基于距离的闭环检测方法、装置、电子设备和存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107796397A (zh) * 2017-09-14 2018-03-13 杭州迦智科技有限公司 一种机器人双目视觉定位方法、装置和存储介质
KR101988555B1 (ko) * 2017-12-05 2019-06-12 충북대학교 산학협력단 조명 불변 영상을 사용한 slam 시스템 및 점구름 지도 생성 방법
CN109709801A (zh) * 2018-12-11 2019-05-03 智灵飞(北京)科技有限公司 一种基于激光雷达的室内无人机定位系统及方法
CN110689622A (zh) * 2019-07-05 2020-01-14 电子科技大学 一种基于点云分割匹配闭环校正的同步定位与构图算法
CN111553844A (zh) * 2020-04-29 2020-08-18 北京百度网讯科技有限公司 用于更新点云的方法及装置
CN112068154A (zh) * 2020-09-14 2020-12-11 中科院软件研究所南京软件技术研究院 一种激光建图定位方法、装置、存储介质及电子设备
CN112325873A (zh) * 2021-01-04 2021-02-05 炬星科技(深圳)有限公司 一种环境地图自主更新方法、设备及计算机可读存储介质

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115200572A (zh) * 2022-09-19 2022-10-18 季华实验室 三维点云地图构建方法、装置、电子设备及存储介质
CN115200572B (zh) * 2022-09-19 2022-12-09 季华实验室 三维点云地图构建方法、装置、电子设备及存储介质
CN115685223A (zh) * 2022-12-15 2023-02-03 深圳市智绘科技有限公司 位置识别方法、装置、电子设备及可读存储介质
CN117576200A (zh) * 2024-01-15 2024-02-20 山东大学 一种长周期移动机器人定位方法、系统、设备及介质
CN117576200B (zh) * 2024-01-15 2024-05-03 山东大学 一种长周期移动机器人定位方法、系统、设备及介质

Also Published As

Publication number Publication date
CN112325873B (zh) 2021-04-06
CN112325873A (zh) 2021-02-05

Similar Documents

Publication Publication Date Title
WO2022143360A1 (zh) 一种环境地图自主更新方法、设备及计算机可读存储介质
CN108827249B (zh) 一种地图构建方法和装置
CN113899363B (zh) 车辆的定位方法、装置及自动驾驶车辆
CN111210475B (zh) 一种地图更新方法及装置
CN111680747B (zh) 用于占据栅格子图的闭环检测的方法和装置
CN113989451B (zh) 高精地图构建方法、装置及电子设备
CN111209978A (zh) 三维视觉重定位方法、装置及计算设备、存储介质
WO2023273169A1 (zh) 一种融合视觉与激光的2.5d地图构建方法
CN113269878A (zh) 一种基于多传感器的建图方法及系统
CN114255323A (zh) 机器人、地图构建方法、装置和可读存储介质
CN115355901A (zh) 一种融合动态目标感知的多机联合建图方法
CN114690769B (zh) 路径规划方法、电子设备及存储介质、计算机程序产品
CN114689036A (zh) 地图更新方法、自动驾驶方法、电子设备及存储介质
CN112556699B (zh) 导航定位方法、装置、电子设备及可读存储介质
CN113932796A (zh) 高精地图车道线生成方法、装置和电子设备
CN113052761B (zh) 激光点云地图融合方法、设备以及计算机可读存储介质
CN112269848A (zh) 一种众包轨迹数据融合方法及装置
CN112446842A (zh) 基于稀疏和稠密深度图图像重建方法、系统、设备及介质
US20230195134A1 (en) Path planning method
CN115619954A (zh) 稀疏语义地图的构建方法、装置、设备及存储介质
CN115937452A (zh) 一种基于分布式计算的地图重建方法与系统
CN115016545A (zh) 无人机着陆的着陆点自主选择方法、设备及介质
CN114299192A (zh) 定位建图的方法、装置、设备和介质
Milijaš et al. When measurements fail: using an interactive SLAM solution to fight bad odometry
CN111708046A (zh) 一种障碍物的平面数据的处理方法、装置、电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21914097

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21914097

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205N DATED 29/08/2023)