WO2022095654A1 - 基于点云的地图校准方法、系统、机器人及云端平台 - Google Patents

基于点云的地图校准方法、系统、机器人及云端平台 Download PDF

Info

Publication number
WO2022095654A1
WO2022095654A1 PCT/CN2021/122363 CN2021122363W WO2022095654A1 WO 2022095654 A1 WO2022095654 A1 WO 2022095654A1 CN 2021122363 W CN2021122363 W CN 2021122363W WO 2022095654 A1 WO2022095654 A1 WO 2022095654A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
obstacle
map
robot
cloud platform
Prior art date
Application number
PCT/CN2021/122363
Other languages
English (en)
French (fr)
Inventor
孟祥宇
马世奎
董文锋
Original Assignee
达闼机器人股份有限公司
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 达闼机器人股份有限公司 filed Critical 达闼机器人股份有限公司
Priority to JP2021578196A priority Critical patent/JP7465290B2/ja
Priority to US17/563,792 priority patent/US20220147049A1/en
Publication of WO2022095654A1 publication Critical patent/WO2022095654A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
    • G05D1/0251—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means extracting 3D information from a plurality of images taken from different locations, e.g. stereo vision
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • the present disclosure relates to the field of point cloud technology, and in particular, to a point cloud-based map calibration method, system, robot and cloud platform.
  • the digital twin virtual-reality combined robot has functions such as planning routes and executing operations.
  • the digital map information pre-made by the robot cannot be real-time. Reflect the real environment at that time, such as dynamic environmental obstacles (tables, chairs, people, etc.), environmental changes (room furniture layout changes, etc.) and so on.
  • operations such as robot navigation and grasping, it is easy to collide with obstacles, the success rate of navigation or grasping is reduced, and the risk factor of robot action is increased.
  • the embodiments of the present disclosure provide a point cloud-based map calibration method, system, robot, and cloud platform, which can calibrate the map information of the robot through three-dimensional point cloud reconstruction.
  • an embodiment of the present disclosure provides a point cloud-based map calibration method.
  • the method is applied to a cloud platform, and the cloud platform is in communication connection with a designated robot, including: obtaining environment collection information from the designated robot; Collect information to reconstruct a 3D point cloud, perform obstacle identification on the 3D point cloud reconstruction result, and obtain an obstacle identification result; the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information; when it is determined as When the confidence information satisfies the first preset index, the map calibration information corresponding to the obstacle information is sent to the designated robot, so that the designated robot calibrates the map information according to the map calibration information.
  • the cloud platform is also communicatively connected with a supervisory terminal, and the method further includes: when it is determined that the confidence information does not meet the first preset index, sending an obstacle confirmation request to the supervisory terminal to Requesting the supervisory terminal to determine whether the obstacle information corresponding to the confidence information satisfies the second preset index; receiving obstacle feedback information from the supervisory terminal, the obstacle feedback information carrying the information corresponding to the obstacle information A determination result; when the determination result is that the obstacle information corresponding to the confidence information satisfies the second preset index, send the map calibration information corresponding to the obstacle information to the designated robot, so that all The designated robot calibrates the map information according to the map calibration information.
  • the method further includes: receiving instruction planning information from the designated robot, the instruction planning information carrying the first planned path; sending the 3D point cloud reconstruction result and the environment collection information corresponding to the first planned path to the supervisor terminal, so that the supervisor terminal can judge the Whether the first planned path complies with the third preset index to obtain a judgment result; receive the judgment result from the supervisory terminal, and send planning feedback information to the designated robot based on the judgment result, so that the designated robot can be based on the The planning feedback information determines a second planning path; wherein the first planning path and the second planning path are the same or different.
  • Another aspect of an embodiment of the present disclosure provides a point cloud-based map calibration method, the method is applied to a robot, and the robot is communicatively connected to a cloud platform, and the method includes: sending environment collection information to the cloud platform, so that all The cloud platform performs three-dimensional point cloud reconstruction according to the environment collection information, performs obstacle identification on the three-dimensional point cloud reconstruction result, and obtains an obstacle identification result; the obstacle identification result includes obstacle information and the corresponding obstacle information. confidence information; receive map calibration information from the cloud platform, where the map calibration information corresponds to obstacle information whose confidence information meets a first preset index; calibrate the map information according to the map calibration information, Get calibration map information.
  • the method before sending the environment collection information to the cloud platform, the method further includes: obtaining a planning instruction, where the planning instruction is used to instruct to perform path planning on the map information according to a specified operation; The planning instruction performs information collection to obtain environment collection information.
  • the method further includes: performing path planning for the specified operation on the calibrated map information, to obtain the first map information. planning a path; sending the first planning path to the cloud platform to instruct the cloud platform to judge whether the first planning path meets the third preset index through the supervisor terminal to obtain a judgment result; receiving planning feedback information from the cloud platform, The planning feedback information corresponds to the judgment result, and a second planning path is determined based on the planning feedback information; wherein the first planning path and the second planning path are the same or different.
  • a cloud platform is connected to a designated robot in communication, and includes: a first obtaining module for obtaining environment collection information from the designated robot; and a reconstruction module for collecting information on the environment performing three-dimensional point cloud reconstruction, performing obstacle identification on the three-dimensional point cloud reconstruction result, and obtaining an obstacle identification result;
  • the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information;
  • the first sending module When it is determined that the confidence information satisfies the first preset index, send the map calibration information corresponding to the obstacle information to the designated robot, so that the designated robot can adjust the calibration information according to the map calibration information. map information for calibration.
  • the cloud platform is also communicatively connected to a supervisory terminal, and further includes: the first sending module, further configured to send the obstacle when it is determined that the confidence information does not meet the first preset index.
  • a confirmation request is sent to the supervisory terminal to request the supervisory terminal to determine whether the obstacle information corresponding to the confidence information meets the second preset index; the first receiving module is used for receiving the obstacle feedback information from the supervisory terminal.
  • the obstacle feedback information carries a determination result corresponding to the obstacle information; the first sending module is further used for when the determination result is that the obstacle information corresponding to the confidence information satisfies the second preset index At the time of sending, map calibration information corresponding to the obstacle information is sent to the designated robot, so that the designated robot calibrates the map information according to the map calibration information.
  • the method further includes: the first receiving module, further configured to receive instruction planning information from the designated robot, where the instruction planning information carries a first planning path; the first sending module , and is also used to send the 3D point cloud reconstruction result and the environment collection information corresponding to the first planned path to the supervisor terminal, so that the supervisor terminal can judge the first planned path based on the 3D point cloud reconstruction result and the environmental collection information Whether it meets the third preset index to obtain the judgment result; the first receiving module is further configured to receive the judgment result from the supervision terminal, and send planning feedback information to the designated robot based on the judgment result, so that all The designated robot determines a second planned path based on the planning feedback information; wherein the first planned path and the second planned path are the same or different.
  • a robot the robot is in communication connection with a cloud platform, and includes: a second sending module, configured to send environment collection information to the cloud platform, so that the cloud platform can respond to the environment according to the environment.
  • a second sending module configured to send environment collection information to the cloud platform, so that the cloud platform can respond to the environment according to the environment.
  • the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information
  • the second receiving a module for receiving map calibration information from the cloud platform, where the map calibration information corresponds to the obstacle information whose confidence information satisfies a first preset index; a calibration module for pairing according to the map calibration information
  • the map information is calibrated to obtain the calibrated map information.
  • a second obtaining module configured to obtain a planning instruction, where the planning instruction is used to instruct to perform path planning on the map information according to a specified operation
  • a collection module configured to perform path planning based on the planning Instructions to collect information and obtain environmental collection information.
  • it further includes: a planning module, configured to perform path planning for the specified operation on the calibrated map information, to obtain a first planned path; the second sending module, further used sending the first planned path to the cloud platform to instruct the cloud platform to judge whether the first planned path conforms to the third preset index through the supervisor terminal to obtain a judgment result; the second receiving module is also used for receiving Planning feedback information from the cloud platform, where the planning feedback information corresponds to the judgment result, and a second planning path is determined based on the planning feedback information; wherein the first planning path and the second planning path are the same or different .
  • a point cloud-based map calibration system the system includes a cloud platform and a designated robot, the cloud platform is in communication connection with the designated robot, and the cloud platform includes: a first obtaining module, which uses to obtain the environmental acquisition information from the designated robot; the reconstruction module is used to reconstruct the 3D point cloud of the environmental acquisition information, and to identify obstacles on the 3D point cloud reconstruction results to obtain the obstacle identification results; the obstacle identification results include: obstacle information and confidence information corresponding to the obstacle information; a first sending module, configured to calibrate the map corresponding to the obstacle information when it is determined that the confidence information meets the first preset index The information is sent to the designated robot, so that the designated robot calibrates the map information according to the map calibration information.
  • the calibration method provided by the embodiment of the present disclosure is used to obtain obstacle information and corresponding confidence information in a real environment according to three-dimensional point cloud reconstruction, and to calibrate the map information of the robot according to the map calibration information corresponding to the obstacle information, so that the robot It can better integrate and calibrate the obstacles in the real environment and map information, so that when the robot navigates or performs other operations according to the map information, it can avoid collision with obstacles in the real environment, reduce the risk factor of robot actions, and improve Safety of robot movements.
  • FIG. 1 is a schematic diagram of an implementation flow of a point cloud-based map calibration method according to an embodiment of the present disclosure
  • FIG. 2 is a schematic flowchart of the implementation of a point cloud-based map calibration method for a supervisor to judge obstacle information according to an embodiment of the present disclosure
  • FIG. 3 is a schematic flowchart of an implementation of a point cloud-based map calibration method for calibrating a planned path according to an embodiment of the present disclosure
  • FIG. 4 is a schematic flowchart of scene implementation of a point cloud-based map calibration method according to an embodiment of the present disclosure
  • FIG. 5 is a schematic diagram of a scene of a point cloud-based map calibration method according to an embodiment of the present disclosure
  • FIG. 6 is a scene rendering diagram after a point cloud-based map calibration method is executed according to an embodiment of the present disclosure
  • FIG. 7 is a schematic diagram of implementation modules of a cloud platform according to an embodiment of the present disclosure.
  • FIG. 1 is a schematic diagram of an implementation flow of a point cloud-based map calibration method according to an embodiment of the present disclosure.
  • an embodiment of the present disclosure provides, on the one hand, a point cloud-based map calibration method.
  • the method is applied to a cloud platform, and the cloud platform is communicatively connected to a designated robot, including: step 101 , obtaining environment collection information from the designated robot; step 102. Perform three-dimensional point cloud reconstruction on the collected environmental information, and perform obstacle identification on the three-dimensional point cloud reconstruction result to obtain an obstacle identification result; the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information; step 103, when it is determined When the confidence information meets the first preset index, the map calibration information corresponding to the obstacle information is sent to the designated robot, so that the designated robot calibrates the map information according to the map calibration information.
  • the calibration method provided by the embodiment of the present disclosure is used to obtain obstacle information and corresponding confidence information in a real environment according to three-dimensional point cloud reconstruction, and when the confidence information meets the first preset index, according to the map corresponding to the obstacle information
  • the calibration information calibrates the map information of the robot, so that the robot can better integrate and calibrate the real environment and obstacles in the map information, so that when the robot navigates or performs other operations according to the map information, it can avoid conflicts with the real environment. Obstacles collide, reduce the risk factor of the robot's action, and improve the safety of the robot's action.
  • the cloud platform in step 101, can establish a communication connection with multiple robots, and the cloud platform records the identity identifier corresponding to each robot, and the identity identifier of each robot is unique, so that the cloud platform can The platform can identify and distinguish multiple robots.
  • the cloud platform obtains the environmental collection information from the designated robot through communication transmission.
  • the designated robot is one of multiple robots that communicate with the cloud platform, and the environment collection information can be used to characterize that the designated robot collects information corresponding to the real environment through radar and/or cameras, such as photos corresponding to the real environment or radar images.
  • the cloud platform includes an information receiving module and a cloud computer vision module.
  • the information receiving module of the cloud platform receives the environment collection information, it sends the environment collection information to the cloud computer vision module, and the cloud computer vision module sends the environment collection information to the cloud computer vision module.
  • Real-time 3D point cloud reconstruction is performed on the environmental acquisition information to reconstruct the environmental acquisition information into 3D point cloud reconstruction results.
  • the cloud computer vision module performs visual recognition on the reconstructed 3D point cloud reconstruction results to identify obstacles in the real environment.
  • the obstacle identification result corresponding to the obstacle wherein the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information.
  • the obstacle information is used to represent attribute parameters corresponding to obstacles in the real environment, including but not limited to at least one of the following: obstacle type, obstacle size, obstacle position, obstacle material, etc.; further, the attribute parameter is Obstacle Type, Obstacle Size, and Obstacle Location.
  • the cloud computer vision module can identify the 2D and 3D bounding boxes corresponding to obstacles in the real environment, and generate the obstacle type and confidence information corresponding to the 2D and 3D bounding boxes by marking. Confidence information is used to evaluate the confidence of obstacle information. It can be understood that the 3D point cloud reconstruction result includes at least one piece of obstacle information, usually multiple pieces, and each obstacle information corresponds to a piece of confidence information. And the confidence information corresponding to different obstacle information may be the same or different.
  • the first preset index is used to evaluate the confidence information, and the first preset index may also be one or more.
  • the confidence information satisfies the first preset index
  • Generate map calibration information according to the obstacle information corresponding to the confidence information and send the map calibration information to the designated robot.
  • the obstacle information corresponding to the confidence information can be the 3D point cloud reconstruction result.
  • the 3D point cloud reconstruction result corresponding to the obstacle information needs to be converted into a map calibration.
  • the information, specifically, the obstacle information can be represented by the three-dimensional point cloud reconstruction result, and the map calibration information can be represented by the map information corresponding to the map.
  • the map information also includes, but is not limited to, obstacle type, obstacle size, obstacle location, obstacle material, and other information.
  • the designated robot does not need to set up function modules related to the point cloud, and the map information can be directly calibrated by receiving the map calibration information from the cloud platform.
  • the confidence information can be characterized by a confidence value
  • the corresponding first preset index can be characterized by setting a confidence threshold.
  • the information of the obstacle information is trusted.
  • the first preset index may be set according to a certain attribute corresponding to the obstacle information, that is, different obstacle information may be matched with different first preset indexes according to their corresponding attributes.
  • map information can be created based on the calibration information from the map, and in subsequent use, the map information can be calibrated through the map calibration information. It is understandable that when the map calibration information calibrates the map information, the designated robot will compare the map calibration information with the obstacles in the map information. to be calibrated.
  • the calibration of the map includes but is not limited to deleting obstacles and setting obstacles. For example, if there is obstacle A in the map calibration information, but there is no obstacle A in the map information, then the map calibration information corresponding to the obstacle A is in the map information. Obstacle A is set; if there is no obstacle B in the map calibration information, but there is an obstacle B in the map information, then according to the map calibration information corresponding to the obstacle B, the obstacle B is deleted in the map information.
  • the cloud platform can send map calibration information to the designated robot multiple times, and each time the information sent may correspond to one or more obstacles.
  • the designated robot calibrates the map information in batches according to the received map calibration information. In this case, when the designated robot deletes obstacles on the map information, it needs to receive all the obstacle information before proceeding.
  • the robot is set up with radar and RGBD cameras.
  • the cloud robot control platform includes a computer vision module.
  • the method includes:
  • Operation 1 The robot uploads the information captured by the radar and RGBD camera to the cloud computer vision module of the cloud robot control platform (HARI) in real time;
  • HARI cloud robot control platform
  • Operation 2 The cloud computer vision module reconstructs the real-time 3D point cloud of the information uploaded by the robot;
  • the cloud computer vision module identifies the reconstructed 3D point cloud, identifies obstacles and 2D and 3D bounding boxes in the current environment, and marks the type and confidence of the current obstacle;
  • Operation 4 The cloud computer vision module sends the real-time 3D point cloud and recognition results to the cloud robot control platform through the cloud platform;
  • Operation 5 The cloud robot control platform converts the successfully marked obstacles into a format adapted to the map information and sends it to the robot to instruct the robot to calibrate the map information according to the obstacles in the format adapted to the map information, and obtain the calibration After the map information.
  • FIG. 2 is a schematic diagram of an implementation flow of a point cloud-based map calibration method for a supervisor to judge obstacle information according to an embodiment of the present disclosure.
  • the cloud platform is also communicatively connected to a supervisory terminal, and the method further includes: Step 201, when it is determined that the confidence information does not meet the first preset index, send an obstacle confirmation request to the supervisory terminal, To request the supervisory terminal to determine whether the obstacle information corresponding to the confidence information satisfies the second preset index; Step 202, receive obstacle feedback information from the supervisory terminal, and the obstacle feedback information carries a determination result corresponding to the obstacle information; Step 203, When the determined result is that the obstacle information corresponding to the confidence information satisfies the second preset index, the map calibration information corresponding to the obstacle information is sent to the designated robot, so that the designated robot calibrates the map information according to the map calibration information.
  • the supervision terminal can be a management user terminal that communicates with the cloud platform.
  • the supervision terminal can be operated by the staff.
  • the staff judges the obstacle information through the supervision terminal to determine whether the designated robot needs to calibrate the map information according to the obstacle information. , and the specific operation content of calibrating the map information according to the obstacle information.
  • the cloud platform sends several of the environmental collection information, reconstructed point cloud, obstacle information, and confidence information to the supervision terminal in the obstacle confirmation request, so that the supervision terminal determines the corresponding confidence information according to the above information. Whether the obstacle information of satisfies the second preset index.
  • the obstacle confirmation request from the cloud platform only carries the obstacle information and the 3D point cloud reconstruction result corresponding to the obstacle information, and the supervisor determines whether the obstacle information satisfies the second prediction according to the 3D point cloud reconstruction result.
  • An index is set, and the second preset index is used to evaluate whether the obstacle information corresponds to an obstacle in the real environment.
  • the obstacle feedback information is sent to the cloud platform to instruct the cloud platform that the obstacle information corresponds to an obstacle in the real environment. It is understandable that when the supervisory terminal staff determines that the obstacle information does not correspond to the obstacles in the real environment, the supervisory terminal also sends the obstacle feedback information to the cloud platform to instruct the cloud platform that the obstacle information does not correspond to the obstacles in the real environment. obstacles.
  • step 202 of the embodiment of the present disclosure after receiving the obstacle feedback information from the supervisor, the cloud platform can know the corresponding determination result by analyzing the obstacle feedback information, and the determination result includes that the obstacle information does not correspond to the obstacle in the real environment The object and obstacle information correspond to the obstacles in the real environment, these two results.
  • step 203 of the embodiment of the present disclosure when the determination result is that the obstacle information corresponding to the confidence information satisfies the second preset index, that is, when the obstacle information corresponds to an obstacle in the real environment, the cloud platform stores the obstacle information Convert it to the corresponding map format and send it to the designated robot, that is, send the map calibration information to the designated robot, so that the designated robot can calibrate the map information according to the map calibration information.
  • the cloud platform can also send obstacle feedback information to the designated robot to notify the designated robot that the obstacles in the real environment have been confirmed. Delete the obstacles that do not correspond to the map calibration information.
  • FIG. 3 is a schematic diagram of an implementation flowchart of a point cloud-based map calibration method for calibrating a planned path according to an embodiment of the present disclosure.
  • the method further includes: Step 301, receiving instruction planning information from the designated robot, where the instruction planning information carries The first planned path; Step 302, send the 3D point cloud reconstruction result and the environment collection information corresponding to the first planned path to the supervision terminal, so that the supervision terminal can judge whether the first planned path conforms to the first planned path based on the 3D point cloud reconstruction result and the environmental collection information
  • the third preset index is to obtain the judgment result; Step 303, receive the judgment result from the supervisor, send planning feedback information to the designated robot based on the judgment result, so that the designated robot determines the second planning path based on the planning feedback information; wherein, the first The planned path is the same or different from the second planned path.
  • the trigger condition for the designated robot to collect information through the radar and the RGBD camera is that the designated robot obtains a planning instruction, and the planning instruction is used to instruct the designated robot to perform path planning.
  • the information collected by the specified robot radar and RGBD camera is specifically the direction pointed by the planning purpose in the planning instruction. For example, if the planning instruction is to go to the position A in the northwest corner of the room, the designated robot will collect the environmental collection information towards the position A in the northwest corner of the room through radar and RGBD cameras.
  • the designated robot After the designated robot obtains the map calibration information from the cloud platform according to the environment collection information, and after the designated robot calibrates the map information according to the map calibration information, the designated robot will plan the first planning path corresponding to the planning instruction according to the calibrated map information. After the designated robot completes the first planned path, the instruction planning information including the first planned path is generated, and the instruction planning information is sent to the cloud platform.
  • the cloud platform may analyze the instruction planning information to determine whether the first planning path can be used to designate the robot to complete the operation corresponding to the planning instruction.
  • the first planning path includes a walking path and an operation path.
  • the walking path is used to represent the movement path of the designated robot in the designated area
  • the operation path is used to represent the joints of the designated robot when performing the operation corresponding to the planning instruction. movement path, etc.
  • the first planned route is sent to the cloud platform in the form of a map.
  • the cloud platform sends the 3D point cloud reconstruction result and the environment collection information corresponding to the first planned path to the supervision terminal, so that the supervision terminal can judge the first plan based on the 3D point cloud reconstruction result and the environment collection information Whether a planned path complies with the third preset index is obtained to obtain a judgment result.
  • the third preset index is used to indicate whether the reconstruction result of the 3D point cloud corresponding to the first planned path is consistent with the environmental collection information.
  • the consistency includes at least the consistency of size and position.
  • the cloud platform receives the judgment result from the supervisor.
  • the judgment result may be a judgment result indicating that the reconstruction result of the 3D point cloud corresponding to the first planned path is consistent with the environmental collection information, or it may be a judgment result that indicates that the corresponding first planned path is consistent with the environmental collection information.
  • the cloud platform sends planning feedback information to the designated robot to indicate that the designated robot can determine the first planned path as the second planned path, and The specified operation is performed according to the second planned path, in this case, the first planned path is the same as the second planned path.
  • the judgment result of the supervisor terminal also carries the path adjustment information generated based on the point cloud, and the cloud platform will adjust the path adjustment information based on the point cloud generation. Convert to map-based path adjustment information, and send planning feedback to the designated robot.
  • the planning feedback information carries the path adjustment information generated based on the map, to instruct the designated robot to adjust the first planned path based on the path adjustment information generated by the map, obtain the second planned path, and execute according to the second planned path Specifies the operation, in which case the first planned path is different from the second planned path.
  • the path adjustment information may be the adjustment of the path or the adjustment of the obstacle. In this method, the path adjustment information is the adjustment of the obstacle.
  • Another aspect of the embodiments of the present disclosure provides a point cloud-based map calibration method, the method is applied to a robot, and the robot is communicated and connected to a cloud platform, and the method includes: first, sending environment collection information to the cloud platform, so that the cloud platform performs a calibration process according to the environment collection information 3D point cloud reconstruction, carry out obstacle recognition on the 3D point cloud reconstruction result, and obtain the obstacle recognition result; then, the obstacle recognition result includes the obstacle information and the confidence information corresponding to the obstacle information; then, receive the map calibration from the cloud platform information, the map calibration information corresponds to the obstacle information whose confidence information satisfies the first preset index; finally, the map information is calibrated according to the map calibration information, and the calibrated map information is obtained.
  • the calibration method provided by the embodiment of the present disclosure is used to obtain obstacle information and confidence information in a real environment according to three-dimensional point cloud reconstruction.
  • the map information of the robot is calibrated according to the obstacle information. Calibration, so that the robot can better integrate and calibrate the obstacles in the real environment and map information, so that when the robot can navigate or perform other operations according to the map information, it can avoid the collision risk with the obstacles in the real environment and reduce the robot.
  • the risk factor of heartbeat improves the safety of robot action.
  • the cloud platform of the embodiment of the present disclosure can establish a communication connection with multiple robots, each robot has a corresponding ID, and different robots have different IDs, so that the cloud platform can distinguish multiple robots.
  • the robot collects environment collection information corresponding to the real environment through radar and/or cameras.
  • the environmental acquisition information may be in the form of pictures, radar images, etc., which are directly acquired by radar and/or cameras, that is, the robot does not perform format conversion of the environmental acquisition information.
  • the cloud platform obtains the environmental collection information from the robot through communication transmission.
  • the cloud platform sends the environmental acquisition information to the cloud computer vision module, and the cloud computer vision module performs real-time 3D point cloud reconstruction on the environmental acquisition information to reconstruct the environmental acquisition information into a 3D point cloud, and the cloud computer vision module reconstructs the reconstructed 3D points.
  • the cloud performs visual recognition to recognize the obstacle recognition result corresponding to the obstacle in the real environment, wherein the obstacle recognition result includes obstacle information and confidence information corresponding to the obstacle information.
  • the obstacle information is used to represent attribute parameters corresponding to obstacles in the real environment, including but not limited to at least one of the following: obstacle type, obstacle size, obstacle position, obstacle material, etc.; further, the attribute parameter is Obstacle Type, Obstacle Size, and Obstacle Location.
  • the cloud computer vision module can identify the 2D and 3D bounding boxes corresponding to obstacles in the real environment, and generate the obstacle type and confidence information corresponding to the 2D and 3D bounding boxes by marking. Confidence information is used to evaluate the confidence of obstacle information. It can be understood that the reconstructed 3D point cloud includes at least one piece of obstacle information, usually multiple pieces, and each obstacle information corresponds to a piece of confidence information.
  • the first preset index is used to evaluate each piece of confidence information separately, and when any piece of confidence information satisfies the first preset index, the obstacle information corresponding to the confidence information is sent to the robot.
  • the obstacle information corresponding to the confidence level can be a point cloud.
  • the point cloud corresponding to the obstacle information needs to be converted into map information. Including but not limited to obstacle type, obstacle size, obstacle location, obstacle material and other information. In this way, the robot does not need to set up function modules related to the point cloud, and the map information can be calibrated by receiving the map calibration information from the cloud platform.
  • the confidence information can be characterized by a confidence value, and the corresponding first preset index can be characterized by setting a confidence threshold. When it is determined that the confidence information meets the first preset index, it can be understood as the corresponding confidence.
  • the information of the obstacle information is trusted.
  • the first preset index may be set according to the type in the obstacle information, that is, the first preset index corresponding to different obstacle types may be different. It should be further added that when the robot is used for the first time, map information can be created based on the obstacle information from the cloud platform, and in subsequent use, the map information can be calibrated based on the map calibration information from the cloud platform.
  • the robot when calibrating the map information based on the map calibration information from the cloud platform, the robot will compare whether the map calibration information is consistent with the obstacles in the map information.
  • the corresponding map information is calibrated.
  • the calibration of the map includes but is not limited to deleting obstacles and setting obstacles. For example, if there is obstacle A in the map calibration information, but there is no obstacle A in the map information, then the map calibration information corresponding to the obstacle A is in the map information. Obstacle A is set; if there is no obstacle B in the map calibration information, but there is an obstacle B in the map information, then according to the map calibration information corresponding to the obstacle B, the obstacle B is deleted in the map information. Further, the actual time-consuming of evaluating each confidence information based on the cloud platform may be different.
  • the cloud platform can send map calibration information to the robot multiple times, and each map calibration information can correspond to one or more obstacle information. Send, and the robot calibrates the map information in batches according to the received map calibration information. In this case, when the robot deletes obstacles on the map information, it needs to receive all the map calibration information before doing it.
  • the method before sending the environment collection information to the cloud platform, the method further includes: first, obtaining a planning instruction, where the planning instruction is used to instruct to perform path planning on the map information according to a specified operation; then, performing information based on the planning instruction. Collection, to obtain environmental collection information.
  • the trigger condition for the robot to collect information through the radar and the RGBD camera is that the robot obtains a planning instruction, and the planning instruction is used to instruct the robot to perform path planning.
  • the information collected by the robot radar and the RGBD camera is specifically the direction pointed by the planning purpose in the planning instruction. For example, if the planning instruction is to go to the position A in the northwest corner of the room, the robot will collect the environment collection information towards the position A in the northwest corner of the room through radar and RGBD cameras.
  • the method further includes: first, performing path planning for the designated operation on the calibrated map information to obtain a first planned path; then, sending the first planned path. Plan the route to the cloud platform to instruct the cloud platform to judge whether the first planned route meets the third preset index through the supervisor to obtain the judgment result; then, receive the planning feedback information from the cloud platform, and the planning feedback information corresponds to the judgment result , determining a second planned path based on the planning feedback information; wherein the first planned path is the same as or different from the second planned path.
  • the robot After the robot obtains the map calibration information from the cloud platform according to the environment collection information, the robot calibrates the map information according to the map calibration information.
  • the designated robot will perform the first planning path corresponding to the planning instruction according to the calibrated map information.
  • the robot After the robot completes the first planned path, it generates instruction planning information including the first planned path, and sends the instruction planning information to the cloud platform.
  • the cloud platform may analyze the instruction planning information to determine whether the first planned path can be used for the robot to complete the operation corresponding to the planning instruction.
  • the first planned path includes a walking path and an operation path.
  • the walking path is used to represent the movement path of the robot in the designated area
  • the operation path is used to represent the joint movement when the robot performs the operation corresponding to the planning instruction. path etc.
  • the cloud platform After receiving the instruction planning information, the cloud platform sends the 3D point cloud reconstruction result and environment collection information corresponding to the first planned path to the supervision terminal, so that the supervision terminal can judge whether the first planned path is based on the 3D point cloud reconstruction result and the environment collection information.
  • the third preset index is met to obtain the judgment result.
  • the third preset index is used to represent whether the reconstruction result of the 3D point cloud corresponding to the first planned path is consistent with the environmental collection information.
  • the supervisor terminal After generating the corresponding judgment result, the supervisor terminal sends the judgment result to the cloud platform.
  • the cloud platform receives the judgment result from the supervision terminal, and the judgment result may be the judgment result that the 3D point cloud reconstruction result corresponding to the first planned path is consistent with the environmental collection information, or the 3D point cloud reconstruction result representing the corresponding first planned path and Judgment results of inconsistent environmental collection information.
  • the cloud platform sends planning feedback information to the robot to instruct the robot to determine the first planned path as the second planned path, and according to the first planned path
  • the second planned path performs the specified operation, in this case, the first planned path is the same as the second planned path.
  • the judgment result of the supervisor terminal also carries the path adjustment information generated based on the point cloud, and the cloud platform will adjust the path adjustment information based on the point cloud generation. Convert to map-based path adjustment information and send planning feedback to the robot.
  • the planning feedback information carries the path adjustment information generated based on the map to instruct the robot to perform the first planned path based on the path adjustment information generated by the map, obtain the second planned path, and perform the specified operation according to the second planned path , in this case, the first planned path is different from the second planned path.
  • the path adjustment information may be the adjustment of the path or the adjustment of the obstacle. In this method, the path adjustment information is the adjustment of the obstacle.
  • FIG. 4 is a schematic diagram of a scene implementation flow of a point cloud-based map calibration method according to an embodiment of the present disclosure
  • FIG. 5 is a scene schematic diagram of a point cloud-based map calibration method according to an embodiment of the present disclosure
  • FIG. 6 is the first embodiment of the present disclosure. A scene rendering after the execution of a point cloud-based map calibration method.
  • the map calibration methods include:
  • Step 401 the robot receives a planning instruction from the user, and the planning instruction is used to instruct the robot to grab the water cup placed on the desktop;
  • Step 402 the robot follows the direction pointed by the water cup, captures the environment collection information through the radar and the RGBD camera and uploads it to the computer vision module of the cloud robot control platform (HARI);
  • HARI cloud robot control platform
  • Step 403 the cloud computer vision module performs real-time 3D point cloud reconstruction on the environment collection information uploaded by the robot, and obtains a real-time 3D point cloud reconstruction result;
  • Step 404 the cloud computer vision module identifies the reconstructed real-time 3D point cloud reconstruction result, obtains the recognition result, and the recognition result includes the real-time obstacle and the 2D and 3D bounding boxes in the 3D point cloud reconstruction result, and marks the real-time obstacle type and confidence of the object.
  • Step 405 the cloud computer vision module sends the real-time 3D point cloud reconstruction result and all the recognition results to the remote monitoring terminal.
  • Step 406 the cloud computer vision module sends the real-time 3D point cloud reconstruction result and recognition result to HARI through the cloud platform.
  • Step 407 HARI judges the confidence of the obstacle identified by the computer vision module. If the confidence corresponding to the obstacle is higher than 0.9 (the value range is 0-1), the cloud robot control platform marks the obstacle as existing in the real environment and send the map calibration information corresponding to the obstacle to the robot, so that the robot can calibrate the obstacle in the three-dimensional map carried by the robot.
  • Step 408 when the confidence level corresponding to the obstacle is not higher than 0.9, then HARI sends the obstacle confirmation request to the remote supervisory end, the remote supervisory end is controlled by the human operator, after the remote supervisory end receives the obstacle confirmation request, it will The reconstruction result of the 3D point cloud corresponding to the obstacle is rendered into the 3D map to determine whether the obstacle is an obstacle existing in the real environment.
  • Step 409 the remote supervision terminal sends the result of determining whether the obstacle is an obstacle existing in the real environment to HARI, and HARI determines whether it is necessary to send the corresponding map calibration information to the robot according to the result, so that the robot is in the three-dimensional map carried by the robot. Calibrate the obstacle.
  • Step 410 after the robot completes the calibration of the three-dimensional map, it performs path planning according to the planning instruction, obtains a first planned path, and sends the first planned path to HARI.
  • Step 411 after the HARI is converted into a 3D point cloud reconstruction result and collection information according to the first planned path, the 3D point cloud reconstruction result and collection information are sent to the remote monitoring terminal.
  • Step 412 the remote supervision terminal renders the reconstruction result of the 3D point cloud corresponding to the obstacle into the 3D map, and generates a 3D object model of the corresponding type and size according to the size of the obstacle and the 2D and 3D bounding boxes in the recognition result, and places it on the 3D object model. the corresponding position on the 3D map.
  • Step 413 by comparing the matching degree of the 3D point cloud reconstruction result rendered into the 3D map and the 3D object model, the type and size of the object that the robot can grasp in the real environment can be determined, and the 3D point reconstruction result and the 3D object can be determined according to the matching degree.
  • the degree of matching of the model, the human operator can choose to adjust and mark the object, and the adjustment and calibration of the object include but are not limited to type, size, position, and rotation. For example, it is found that cup 1 has not been recognized, and the size of the recognition result of cup 2 does not match the real size. As a result, when grabbing cup 2, the robot's hand may collide with cup 1, causing a safety risk. Therefore, the human operator adds cup 1 in the virtual environment, and adjusts the size of cup 2 in the virtual environment, so that the calibration of the virtual and real environments is consistent through the point cloud.
  • Step 414 the remote monitoring terminal sends the adjustment and calibration information to HARI, HARI converts the adjustment and calibration information into path adjustment information based on the electronic map, and sends the path adjustment information to the robot, and the robot side re-plans the path according to the received path adjustment information,
  • the second planned path is obtained to guide subsequent navigation.
  • the robot grabs cup 2 it will consider the obstacle avoidance motion plan of cup 1.
  • Step 415 the robot executes the second planned path, and completes the specified operation corresponding to the planning instruction, that is, the robot executes the grasping behavior safely.
  • FIG. 7 is a schematic diagram of an implementation module of a point cloud-based map calibration system according to an embodiment of the present disclosure.
  • FIG. 7 another embodiment of the present disclosure provides a cloud platform, where the cloud platform is communicatively connected to a designated robot, including: a first obtaining module 701 for obtaining environment collection information from the designated robot; a reconstruction module 702 for Perform 3D point cloud reconstruction on the environmental collection information, and perform obstacle identification on the 3D point cloud reconstruction result to obtain an obstacle identification result; the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information; the first sending module 703, using When it is determined that the confidence information meets the first preset index, the map calibration information corresponding to the obstacle information is sent to the designated robot, so that the designated robot calibrates the map information according to the map calibration information.
  • a first obtaining module 701 for obtaining environment collection information from the designated robot
  • a reconstruction module 702 for Perform 3D point cloud reconstruction on the environmental collection information, and perform obstacle identification on the 3D point cloud reconstruction result to obtain an obstacle identification result
  • the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information
  • the first sending module 703 using When it is determined that
  • the cloud platform is also communicatively connected with the supervisor terminal, and further includes: a first sending module 703, further configured to send an obstacle confirmation request to the supervisor terminal when it is determined that the confidence information does not meet the first preset index. , to request the supervisory terminal to determine whether the obstacle information corresponding to the confidence information satisfies the second preset index; the first receiving module 704 is used to receive the obstacle feedback information from the supervisory terminal, and the obstacle feedback information carries the information corresponding to the obstacle information.
  • the determination result; the first sending module 703 is further configured to send the map calibration information corresponding to the obstacle information to the designated robot when the determination result is that the obstacle information corresponding to the confidence information satisfies the second preset index, so as to make the designated robot
  • the robot calibrates the map information according to the map calibration information.
  • it further includes: a first receiving module 704, further configured to receive instruction planning information from a designated robot, where the instruction planning information carries a first planning path; a first sending module 703, further configured to send and The 3D point cloud reconstruction result and the environmental collection information corresponding to the first planned path are sent to the supervision terminal, so that the supervision terminal can judge whether the first planned path conforms to the third preset index based on the 3D point cloud reconstruction result and the environmental collection information, so as to obtain the judgment result
  • the first receiving module 704 is also used to receive the judgment result from the supervisory end, and send planning feedback information to the designated robot based on the judgment result, so that the designated robot determines the second planning path based on the planning feedback information; wherein, the first planning path and the first planning path The two planning paths are the same or different.
  • the robot is communicatively connected to a cloud platform, and includes: a second sending module 705, configured to send environment collection information to the cloud platform, so that the cloud platform can perform 3D point cloud reconstruction according to the environment collection information , perform obstacle identification on the 3D point cloud reconstruction result, and obtain the obstacle identification result;
  • the obstacle identification result includes obstacle information and confidence information corresponding to the obstacle information;
  • the second receiving module 706 is used to receive the map calibration information from the cloud platform , the map calibration information corresponds to the obstacle information whose confidence information meets the first preset index;
  • the calibration module 707 is used for calibrating the map information according to the map calibration information to obtain the calibration map information.
  • it further includes: a second obtaining module 708, configured to obtain a planning instruction, where the planning instruction is used to instruct to perform path planning on the map information according to a specified operation; a collection module 709, configured to perform information collection based on the planning instruction to obtain environmental collection information.
  • a planning module 710 configured to perform path planning for the specified operation on the calibrated map information, to obtain a first planned path
  • a second sending module 705 further configured to send the first planned path to the cloud platform to instruct the cloud platform to judge whether the first planned path meets the third preset index through the supervisor to obtain the judgment result
  • the second receiving module 706 is further configured to receive the planning feedback information from the cloud platform, and the planning feedback information Corresponding to the judgment result, a second planned path is determined based on the planning feedback information; wherein the first planned path and the second planned path are the same or different.
  • a point cloud-based map calibration system the system includes a cloud platform and a designated robot, the cloud platform is connected to the designated robot in communication, and the cloud platform includes: a first obtaining module for obtaining data from the designated robot. Environment collection information; reconstruction module, used to reconstruct the 3D point cloud of the environment collection information, identify obstacles on the reconstruction results of the 3D point cloud, and obtain the obstacle identification result; the obstacle identification result includes the obstacle information and the confidence level corresponding to the obstacle information information; a first sending module, configured to send the map calibration information corresponding to the obstacle information to the designated robot when it is determined that the confidence information meets the first preset index, so that the designated robot can perform the map information according to the map calibration information. calibration.
  • first and second are only used for descriptive purposes, and should not be construed as indicating or implying relative importance or implying the number of indicated technical features. Thus, a feature delimited with “first”, “second” may expressly or implicitly include at least one of that feature.
  • plurality means two or more, unless expressly and specifically defined otherwise.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Electromagnetism (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种基于点云的地图校准方法、系统、机器人及云端平台，方法应用于云端平台，云端平台与指定机器人通信连接，包括：获得来自指定机器人的环境采集信息（101）；对环境采集信息进行三维点云重建，对三维点云重建结果进行障碍识别，获得障碍识别结果；障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息（102）；当确定置信度信息满足第一预设指标时，将与障碍物信息对应的地图校准信息发送至指定机器人，以使指定机器人根据地图校准信息对地图信息进行校准（103）。云端平台通过三维点云重建对机器人的地图信息进行校准，以使机器人能够较好地融合与校准现实环境和地图信息中的障碍物。

Description

基于点云的地图校准方法、系统、机器人及云端平台 技术领域
本公开涉及点云技术领域，尤其涉及一种基于点云的地图校准方法、系统、机器人及云端平台。
背景技术
目前，数字孪生虚实结合机器人具有规划路线和执行操作等功能，在数字孪生虚实结合机器人控制系统的实现过程中，因为现实环境内的物体可能会随时变化，机器人预先制作好的数字化地图信息无法实时反映出当时的现实环境，如动态的环境障碍物(桌椅、人等)、环境改变(房间家具布局变化等)等。导致机器人导航与抓取等操作时，容易与障碍物发生碰撞，导航或抓取成功率降低，机器人行动的危险系数升高。
发明内容
本公开实施例提供了一种基于点云的地图校准方法、系统、机器人及云端平台，能够通过三维点云重建对机器人的地图信息进行校准。
本公开实施例一方面提供一种基于点云的地图校准方法，所述方法应用于云端平台，所述云端平台与指定机器人通信连接，包括：获得来自指定机器人的环境采集信息；对所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；当确定为所述置信度信息满足第一预设指标时，将与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
在一可实施方式中，所述云端平台还通信连接有监管端，所述方法还 包括：当确定为所述置信度信息不满足第一预设指标时，发送障碍确认请求至监管端，以请求所述监管端确定与所述置信度信息对应的障碍物信息是否满足第二预设指标；接收来自所述监管端的障碍反馈信息，所述障碍反馈信息携带有与所述障碍物信息对应的确定结果；当所述确定结果为与所述置信度信息对应的障碍物信息满足第二预设指标时，发送与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
在一可实施方式中，在将与所述障碍物信息对应的地图校准信息发送至所述指定机器人之后，所述方法还包括：接收来自所述指定机器人的指令规划信息，所述指令规划信息中携带有第一规划路径；发送与所述第一规划路径对应的三维点云重建结果和环境采集信息至监管端，以使监管端基于所述三维点云重建结果和环境采集信息判断所述第一规划路径是否符合第三预设指标，以获得判断结果；接收来自所述监管端的判断结果，基于所述判断结果发送规划反馈信息至所述指定机器人，以使所述指定机器人基于所述规划反馈信息确定第二规划路径；其中，所述第一规划路径与所述第二规划路径相同或不同。
本公开实施例另一方面提供基于点云的地图校准方法，所述方法应用于机器人，所述机器人与云端平台通信连接，所述方法包括：发送环境采集信息至所述云端平台，以使所述云端平台根据所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；接收来自所述云端平台的地图校准信息，所述地图校准信息与所述置信度信息满足第一预设指标的障碍物信息对应；根据所述地图校准信息对地图信息进行校准,获得校准地图信息。
在一可实施方式中，在发送环境采集信息至所述云端平台之前，所述 方法还包括：获得规划指令，所述规划指令用于指示根据指定操作在所述地图信息上进行路径规划；基于所述规划指令进行信息采集，获得环境采集信息。
在一可实施方式中，在根据所述地图校准信息对地图信息进行校准之后，所述方法还包括：在所述校准后的所述地图信息上对所述指定操作进行路径规划，获得第一规划路径；发送所述第一规划路径至云端平台，以指示云端平台通过监管端判断所述第一规划路径是否符合第三预设指标，以获得判断结果；接收来自云端平台的规划反馈信息，所述规划反馈信息与所述判断结果对应，基于所述规划反馈信息确定第二规划路径；其中，所述第一规划路径与所述第二规划路径相同或不同。
本公开另一方面提供一种云端平台，所述云端平台与指定机器人通信连接，包括：第一获得模块，用于获得来自指定机器人的环境采集信息；重建模块，用于对所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；第一发送模块，用于当确定为所述置信度信息满足第一预设指标时，将与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
在一可实施方式中，所述云端平台还通信连接有监管端，还包括：所述第一发送模块，还用于当确定为所述置信度信息不满足第一预设指标时，发送障碍确认请求至监管端，以请求所述监管端确定与所述置信度信息对应的障碍物信息是否满足第二预设指标；第一接收模块，用于接收来自所述监管端的障碍反馈信息，所述障碍反馈信息携带有与所述障碍物信息对应的确定结果；所述第一发送模块，还用于当所述确定结果为与所述置信度信息对应的障碍物信息满足第二预设指标时，发送与所述障碍物信息对 应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
在一可实施方式中，还包括：所述第一接收模块，还用于接收来自所述指定机器人的指令规划信息，所述指令规划信息中携带有第一规划路径；所述第一发送模块，还用于发送与所述第一规划路径对应的三维点云重建结果和环境采集信息至监管端，以使监管端基于所述三维点云重建结果和环境采集信息判断所述第一规划路径是否符合第三预设指标，以获得判断结果；所述第一接收模块，还用于接收来自所述监管端的判断结果，基于所述判断结果发送规划反馈信息至所述指定机器人，以使所述指定机器人基于所述规划反馈信息确定第二规划路径；其中，所述第一规划路径与所述第二规划路径相同或不同。
本公开实施例另一方面提供一种机器人，所述机器人与云端平台通信连接，包括：第二发送模块，用于发送环境采集信息至所述云端平台，以使所述云端平台根据所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；第二接收模块，用于接收来自所述云端平台的地图校准信息，所述地图校准信息与所述置信度信息满足第一预设指标的障碍物信息对应；校准模块，用于根据所述地图校准信息对地图信息进行校准，获得校准地图信息。
在一可实施方式中，还包括：第二获得模块，用于获得规划指令，所述规划指令用于指示根据指定操作在所述地图信息上进行路径规划；采集模块，用于基于所述规划指令进行信息采集，获得环境采集信息。
在一可实施方式中，还包括：规划模块，用于在所述校准后的所述地图信息上对所述指定操作进行路径规划，获得第一规划路径；所述第二发送模块，还用于发送所述第一规划路径至云端平台，以指示云端平台通过 监管端判断所述第一规划路径是否符合第三预设指标，以获得判断结果；所述第二接收模块，还用于接收来自云端平台的规划反馈信息，所述规划反馈信息与所述判断结果对应，基于所述规划反馈信息确定第二规划路径；其中，所述第一规划路径与所述第二规划路径相同或不同。
本公开实施例另一方面提供一种基于点云的地图校准系统，所述系统包括云端平台和指定机器人，所述云端平台与指定机器人通信连接，所述云端平台包括：第一获得模块，用于获得来自指定机器人的环境采集信息；重建模块，用于对所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；第一发送模块，用于当确定为所述置信度信息满足第一预设指标时，将与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
本公开实施例提供的校准方法用于根据三维点云重建获得现实环境中的障碍物信息和对应置信度信息，根据与障碍物信息对应的地图校准信息对机器人的地图信息进行校准，以使机器人能够较好地融合与校准现实环境和地图信息中的障碍物，进而使机器人根据地图信息进行导航或执行其他操作时，能够避免与现实环境的障碍物发生碰撞，降低机器人行动的危险系数，提高机器人行动的安全性。
附图说明
通过参考附图阅读下文的详细描述，本公开示例性实施方式的上述以及其他目的、特征和优点将变得易于理解。在附图中，以示例性而非限制性的方式示出了本公开的若干实施方式，其中：
在附图中，相同或对应的标号表示相同或对应的部分。
图1为本公开实施例一种基于点云的地图校准方法的实现流程示意图；
图2为本公开实施例一种基于点云的地图校准方法监管端对障碍物信息进行判断的实现流程示意图；
图3为本公开实施例一种基于点云的地图校准方法校准规划路径的实现流程示意图；
图4为本公开实施例一种基于点云的地图校准方法的场景实现流程示意图；
图5为本公开实施例一种基于点云的地图校准方法的场景示意图；
图6为本公开实施例一种基于点云的地图校准方法执行后的场景效果图；
图7为本公开实施例一种云端平台的实现模块示意图。
具体实施方式
为使本公开的目的、特征、优点能够更加的明显和易懂，下面将结合本公开实施例中的附图，对本公开实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本公开一部分实施例，而非全部实施例。基于本公开中的实施例，本领域技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本公开保护的范围。
图1为本公开实施例一种基于点云的地图校准方法的实现流程示意图。
参见图1，本公开实施例一方面提供一种基于点云的地图校准方法，方法应用于云端平台，云端平台与指定机器人通信连接，包括：步骤101，获得来自指定机器人的环境采集信息；步骤102，对环境采集信息进行三维点云重建，对三维点云重建结果进行障碍识别，获得障碍识别结果；障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息；步骤103，当确定为置信度信息满足第一预设指标时，将与障碍物信息对应的地图校准信息发送至指定机器人，以使指定机器人根据地图校准信息对地图信息进行校准。
本公开实施例提供的校准方法用于根据三维点云重建获得现实环境中的障碍物信息和对应的置信度信息，在置信度信息满足第一预设指标时，根据与障碍物信息对应的地图校准信息对机器人的地图信息进行校准，以使机器人能够较好地融合与校准现实环境和地图信息中的障碍物，进而使机器人根据地图信息进行导航或执行其他操作时，能够避免与现实环境的障碍物发生碰撞，降低机器人行动的危险系数，提高机器人行动的安全性。
具体的，本公开的实施例在步骤101中，云端平台可以与多个机器人建立通信连接，且云端平台记录有每个机器人对应的身份标识，每个机器人的身份标识具有唯一性，以使云端平台能够对多个机器人进行识别区分。云端平台通过通信传输获得来自指定机器人的环境采集信息。指定机器人为与云端平台进行通信连接的多个机器人的其中之一，环境采集信息可以用于表征是由指定机器人通过雷达和/或摄像头采集与现实环境对应的信息，如与现实环境对应的照片或雷达影像。
本公开的实施例在步骤102中，云端平台包括信息接收模块和云端计算机视觉模块，云端平台的信息接收模块接收环境采集信息后，将环境采集信息发送至云端计算机视觉模块，通过云端计算机视觉模块对环境采集信息进行实时三维点云重建，以将环境采集信息重建为三维点云重建结果，云端计算机视觉模块对重建出的三维点云重建结果进行视觉识别，以识别出与现实环境中的障碍物对应的障碍识别结果，其中，障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息。障碍物信息用于表征与现实环境中的障碍物对应的属性参数，包括但不限于如下至少之一：障碍物类型、障碍物尺寸、障碍物位置、障碍物材质等；进一步的，属性参数为障碍物类型、障碍物尺寸和障碍物位置。具体的，云端计算机视觉模块可以识别出与现实环境中的障碍物对应的二维和三维包围盒，并通过标记生成对应二维和三维包围盒的障碍物类型和置信度信息。置信度信息用于评 价障碍物信息的置信度。可以理解的是，三维点云重建结果中包括至少一个障碍物信息，通常为多个，每一个障碍物信息均对应有一个置信度信息。且不同的障碍物信息所对应的置信度信息可能相同或不同。
在本公开实施例的步骤103中，第一预设指标用于对置信度信息进行评价，第一预设指标也可以为1个或多个，当置信度信息满足第一预设指标时，根据与该置信度信息对应的障碍物信息生成地图校准信息，并发送地图校准信息至指定机器人。需要补充的是，置信度信息所对应的障碍物信息可以是三维点云重建结果，云端平台将地图校准信息发送至指定机器人前，需要将对应障碍物信息的三维点云重建结果转换为地图校准信息，具体的，障碍物信息可以通过三维点云重建结果进行表示，地图校准信息通过与地图对应的地图信息进行表示。地图信息同样包括但不限于障碍物类型、障碍物尺寸、障碍物位置、障碍物材质等信息。如此操作，指定机器人上无需设置与点云相关的功能模块，通过接收来自云端平台的地图校准信息即可直接对地图信息进行校准。
进一步的，置信度信息可以通过置信度数值进行表征，对应的第一预设指标可以通过设置置信度阈值进行表征，当确定为置信度信息满足第一预设指标时，可以理解为对应置信度信息的障碍物信息是可信任的。第一预设指标可以根据障碍物信息所对应的某一属性进行设定，即不同的障碍物信息根据其所对应的属性，所匹配的第一预设指标可以不同。
当指定机器人初次使用时，可以基于来自地图校准信息创建地图信息，在后续使用时，可以通过地图校准信息对地图信息进行校准。可以理解的是，在地图校准信息对地图信息进行校准时，指定机器人会比较地图校准信息与地图信息中的障碍物是否一致，在不一致的情况下，对地图信息中与该障碍物对应地图内容的进行校准。地图的校准包括但不限于删除障碍物和设置障碍物，例如在地图校准信息存在障碍物A，在地图信息中不存 在障碍物A，则根据与障碍物A对应的地图校准信息在地图信息中设置障碍物A；在地图校准信息中不存在障碍物B，在地图信息中存在障碍物B，则根据与障碍物B对应的地图校准信息在地图信息中删除障碍物B。
进一步的，由于云端平台对每一个置信度信息进行分别评价的实际耗时可能存在不同，云端平台可以发送多次地图校准信息至指定机器人，每次发送的信息中可以对应有一个或多个障碍物信息，指定机器人根据接收到的地图校准信息对地图信息进行分批校准，在该情况下，当指定机器人对地图信息进行删除障碍物操作时，需要在接收所有障碍物信息后再进行。
为方便上述实施例的理解，以下提供一种具体实施场景进行说明。在该场景中，包括有位于现实环境中的机器人和云端机器人控制平台。机器人设置有雷达和RGBD摄像头。云端机器人控制平台包括有计算机视觉模块。
该方法包括：
操作一：机器人实时将雷达和RGBD摄像头捕获的信息上传至云端机器人控制平台(HARI)的云端计算机视觉模块；
操作二：云端计算机视觉模块将机器人上传的信息进行实时三维点云重建；
操作三：云端计算机视觉模块对重建出的三维点云进行识别，识别出当前环境中的障碍物与二维和三维包围盒，并标记当前障碍物的类型和置信度；
操作四：云端计算机视觉模块通过云端平台将实时三维点云与识别结果发送至云端机器人控制平台；
操作五：云端机器人控制平台将成功标记好的障碍物转换为适配地图信息的格式发送至机器人端，用以指示机器人对根据适配地图信息的格式的障碍物对地图信息进行校准，获得校准后的地图信息。
图2为本公开实施例一种基于点云的地图校准方法监管端对障碍物信息进行判断的实现流程示意图。
参见图2，在一可实施方式中，云端平台还通信连接有监管端，方法还包括：步骤201，当确定为置信度信息不满足第一预设指标时，发送障碍确认请求至监管端，以请求监管端确定与置信度信息对应的障碍物信息是否满足第二预设指标；步骤202，接收来自监管端的障碍反馈信息，障碍反馈信息携带有与障碍物信息对应的确定结果；步骤203，当确定结果为与置信度信息对应的障碍物信息满足第二预设指标时，发送与障碍物信息对应的地图校准信息至指定机器人，以使指定机器人根据地图校准信息对地图信息进行校准。
本方法当确定为置信度信息对应的置信度数值不满足第一预设指标所对应的第一指标阈值时，可以认为该置信度信息所对应的障碍物信息不可信任，在该情况下，可以发送障碍确认请求至监管端。监管端可以为与云端平台通信连接的管理用户端，监管端可以由工作人员进行操作，工作人员通过监管端对障碍物信息进行判断，以确定指定机器人是否需要根据障碍物信息对地图信息进行校准，以及根据障碍物信息对地图信息进行校准的具体操作内容。具体的，云端平台在障碍确认请求中携带环境采集信息、重建的点云、障碍物信息和置信度信息中的其中几项发送至监管端，以使监管端根据上述信息确定与置信度信息对应的障碍物信息是否满足第二预设指标。在一种实施方式中，云端平台的障碍确认请求中仅携带有障碍物信息和与障碍物信息对应的三维点云重建结果，监管端根据三维点云重建结果判断障碍物信息是否满足第二预设指标，第二预设指标用于评价障碍物信息是否对应有现实环境中的障碍物。当监管端的工作人员确定障碍物信息对应有现实环境中的障碍物时，发送障碍反馈信息至云端平台，以指示云端平台该障碍物信息对应有现实环境中的障碍物。可以理解的是，当 监管端工作人员确定障碍物信息不对应有现实环境中的障碍物时，监管端同样发送障碍反馈信息至云端平台，以指示云端平台该障碍物信息不对应有现实环境中的障碍物。
在本公开实施例的步骤202中，云端平台接收来自监管端的障碍反馈信息之后，通过对障碍反馈信息进行分析可以知道对应的确定结果，该确定结果包括障碍物信息不对应有现实环境中的障碍物、及障碍物信息对应有现实环境中的障碍物，这两种结果。
在本公开实施例的步骤203中，当确定结果为与置信度信息对应的障碍物信息满足第二预设指标，即障碍物信息对应有现实环境中的障碍物时，云端平台将障碍物信息转换为对应的地图格式发送至指定机器人，即发送地图校准信息至指定机器人，以使指定机器人根据地图校准信息对地图信息进行校准。可以理解的是，当所有的障碍物信息均判断过是否满足第一预设指标和/或第二预设指标后，即当所有的障碍物信息均确认过是否对应有现实环境中的障碍物时，云端平台还可以发送障碍反馈信息至指定该机器人，以通知指定机器人现实环境中的障碍物均已经确认，此时，指定机器人可以比对所有地图校准信息和地图信息，以在地图信息中删除地图校准信息中不对应的障碍物。
图3为本公开实施例一种基于点云的地图校准方法校准规划路径的实现流程示意图。
参见图3，在一可实施方式中，在将与障碍物信息对应的地图校准信息发送至指定机器人之后，方法还包括：步骤301，接收来自指定机器人的指令规划信息，指令规划信息中携带有第一规划路径；步骤302，发送与第一规划路径对应的三维点云重建结果和环境采集信息至监管端，以使监管端基于三维点云重建结果和环境采集信息判断第一规划路径是否符合第三预设指标，以获得判断结果；步骤303，接收来自监管端的判断结果，基于判 断结果发送规划反馈信息至指定机器人，以使指定机器人基于规划反馈信息确定第二规划路径；其中，第一规划路径与第二规划路径相同或不同。
在本公开实施例中，指定机器人通过雷达和RGBD摄像头进行信息采集的触发条件为指定机器人获得规划指令，规划指令用于指示指定机器人进行路径规划。指定机器人雷达和RGBD摄像头进行采集的信息具体为在规划指令中规划目的所指向的方向。例如，规划指令为走到房间西北角A位置，则指定机器人则通过雷达和RGBD摄像头采集朝向房间西北角A位置的环境采集信息。
在指定机器人根据环境采集信息获得来自云端平台的地图校准信息，指定机器人根据地图校准信息对地图信息进行校准之后，指定机器人会根据校准后的地图信息规划与规划指令对应的第一规划路径。在指定机器人完成第一规划路径后，生成包含第一规划路径的指令规划信息，并发送指令规划信息至云端平台。
云端平台接收指令规划信息后，可以对指令规划信息进行分析，以确定第一规划路径是否能够用于指定机器人完成与规划指令对应的操作。需要说明的是，第一规划路径包括行走路径和操作路径，行走路径用于表征指定机器人在指定区域内的移动路径，操作路径用于表征指定该机器人在执行与规划指令对应的操作时的关节移动路径等。第一规划路径以地图形式发送至云端平台。
在本公开实施例的步骤302中，云端平台通过将与第一规划路径对应的三维点云重建结果和环境采集信息至监管端，以使监管端基于三维点云重建结果和环境采集信息判断第一规划路径是否符合第三预设指标，以获得判断结果。第三预设指标用于表征对应第一规划路径的三维点云重建结果和环境采集信息是否一致，该一致至少包括大小一致和位置一致，监管端生成对应的判断结果后，发送判断结果至云端平台。
在本公开实施例的步骤303中，云端平台接收来自监管端的判断结果，判断结果可以是表征对应第一规划路径的三维点云重建结果和环境采集信息一致的判断结果，也可以是表征对应第一规划路径的三维点云重建结果和环境采集信息不一致的判断结果。
当判断结果为对应第一规划路径的三维点云重建结果和环境采集信息一致时，云端平台发送规划反馈信息至指定机器人，以指示指定机器人可以将第一规划路径确定为第二规划路径，并根据第二规划路径执行指定操作，在该情况下，第一规划路径与第二规划路径相同。
当判断结果为对应第一规划路径的三维点云重建结果和环境采集信息不一致时，监管端的判断结果中还携带有基于点云生成的路径调整信息，云端平台将基于点云生成的路径调整信息转换为基于地图生成的路径调整信息，并发送规划反馈信息至指定机器人。该情况下，规划反馈信息中携带有基于地图生成的路径调整信息，以指示指定机器人基于地图生成的路径调整信息对第一规划路径进行调整，获得第二规划路径，并根据第二规划路径执行指定操作，在该情况下，第一规划路径与第二规划路径不同。路径调整信息可以是对路径的调整，也可以是障碍物的调整，在本方法中，路径调整信息为对障碍物的调整。
本公开实施例另一方面提供基于点云的地图校准方法，方法应用于机器人，机器人与云端平台通信连接，方法包括：首先，发送环境采集信息至云端平台，以使云端平台根据环境采集信息进行三维点云重建，对三维点云重建结果进行障碍识别，获得障碍识别结果；然后，障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息；再后，接收来自云端平台的地图校准信息，地图校准信息与置信度信息满足第一预设指标的障碍物信息对应；最后，根据地图校准信息对地图信息进行校准，获得校准后的地图信息。
本公开实施例提供的校准方法用于根据三维点云重建获得现实环境中的障碍物信息和置信度信息，在置信度信息满足第一预设指标时，根据障碍物信息对机器人的地图信息进行校准，以使机器人能够较好的融合与校准现实环境和地图信息中的障碍物，进而使机器人能够根据地图信息进行导航或执行其他操作时，避免与现实环境的障碍物发生碰撞风险，降低机器人心动的危险系数，提高机器人行动的安全性。
具体的，本公开实施例的云端平台可以与多个机器人建立通信连接，每一个机器人都具有对应的身份标识，不同机器人的身份标识不同，以使云端平台能够对多个机器人进行区分。机器人通过雷达和/或摄像头采集与现实环境对应的环境采集信息。环境采集信息可以为图片、雷达影像等由雷达和/或摄像头直接采集获得的信息形式，即机器人不对环境采集信息进行格式上的转换。云端平台通过通信传输获得来自机器人的环境采集信息。
云端平台将环境采集信息发送至云端计算机视觉模块，通过云端计算机视觉模块对环境采集信息进行实时三维点云重建，以将环境采集信息重建为三维点云，云端计算机视觉模块对重建出的三维点云进行视觉识别，以识别出与现实环境中的障碍物对应的障碍识别结果，其中，障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息。障碍物信息用于表征与现实环境中的障碍物对应的属性参数，包括但不限于如下至少之一：障碍物类型、障碍物尺寸、障碍物位置、障碍物材质等；进一步的，属性参数为障碍物类型、障碍物尺寸和障碍物位置。具体的，云端计算机视觉模块可以识别出与现实环境中的障碍物对应的二维和三维包围盒，并通过标记生成对应二维和三维包围盒的障碍物类型和置信度信息。置信度信息用于评价障碍物信息的置信度。可以理解的是，重建的三维点云中包括至少一个障碍物信息，通常为多个，每一个障碍物信息均对应有一个置信度信息。
第一预设指标用于对每一个置信度信息进行分别评价，当任一个置信度信息满足第一预设指标时，将与该置信度信息对应的障碍物信息发送至机器人。需要补充的是，置信度所对应的障碍物信息可以是点云，云端平台将置信度所对应的障碍物信息发送至机器人前，需要将对应障碍物信息的点云转换为地图信息，地图信息包括但不限于障碍物类型、障碍物尺寸、障碍物位置、障碍物材质等信息。如此操作，机器人上无需设置与点云相关的功能模块，通过接收来自云端平台的地图校准信息即可对地图信息进行校准。进一步的，置信度信息可以通过置信度数值进行表征，对应的第一预设指标可以通过设置置信度阈值进行表征，当确定为置信度信息满足第一预设指标时，可以理解为对应置信度信息的障碍物信息是可信任的。进一步的，第一预设指标可以根据障碍物信息中的类型进行设定，即不同的障碍物类型其所对应的第一预设指标可以不同。进一步需要补充的是，当机器人初次使用时，可以基于来自云端平台的障碍物信息创建地图信息，在后续使用时，可以基于来自云端平台的地图校准信息对地图信息进行校准。可以理解的是，在基于来自云端平台中的地图校准信息对地图信息进行校准时，机器人会比较地图校准信息与地图信息中的障碍物是否一致，在不一致的情况下，对与该地图校准信息对应的地图信息进行校准。地图的校准包括但不限于删除障碍物和设置障碍物，例如在地图校准信息存在障碍物A，在地图信息中不存在障碍物A，则根据与障碍物A对应的地图校准信息在地图信息中设置障碍物A；在地图校准信息中不存在障碍物B，在地图信息中存在障碍物B，则根据与障碍物B对应的地图校准信息在地图信息中删除障碍物B。进一步的，基于云端平台对每一个置信度信息进行分别评价的实际耗时可能存在不同，云端平台可以发送多次地图校准信息至机器人，每次地图校准信息可以对应有一个或多个障碍物信息进行发送，机器人根据接收到的地图校准信息对地图信息进行分批校准，在该情 况下，机器人对地图信息进行删除障碍物操作时，需要在接收所有地图校准信息后再进行。
在一可实施方式中，在发送环境采集信息至云端平台之前，方法还包括：首先，获得规划指令，规划指令用于指示根据指定操作在地图信息上进行路径规划；然后，基于规划指令进行信息采集，获得环境采集信息。
在本公开实施例中，机器人通过雷达和RGBD摄像头进行信息采集的触发条件为机器人获得规划指令，规划指令用于指示机器人进行路径规划。机器人雷达和RGBD摄像头进行采集的信息具体为在规划指令中规划目的所指向的方向。例如，规划指令为走到房间西北角A位置，则机器人则通过雷达和RGBD摄像头采集朝向房间西北角A位置的环境采集信息。
在一可实施方式中，在根据地图校准信息对地图信息进行校准之后，方法还包括：首先，在校准后的地图信息上对指定操作进行路径规划，获得第一规划路径；然后，发送第一规划路径至云端平台，以指示云端平台通过监管端判断第一规划路径是否符合第三预设指标，以获得判断结果；再后，接收来自云端平台的规划反馈信息，规划反馈信息与判断结果对应，基于规划反馈信息确定第二规划路径；其中，第一规划路径与第二规划路径相同或不同。
在机器人根据环境采集信息获得来自云端平台的地图校准信息之后，机器人根据地图校准信息对地图信息进行校准。指定机器人会根据校准后的地图信息进行与规划指令对应的第一规划路径。在机器人完成第一规划路径后，生成包含第一规划路径的指令规划信息，并发送指令规划信息至云端平台。云端平台接收指令规划信息后，可以对指令规划信息进行分析，以确定第一规划路径是否能够用于机器人完成与规划指令对应的操作。需要说明的是，第一规划路径包括行走路径和操作路径，行走路径用于表征机器人在指定区域内的移动路径，操作路径用于表征指定该机器人在执行 与规划指令对应的操作时的关节移动路径等。
云端平台接收指令规划信息之后，通过将与第一规划路径对应的三维点云重建结果和环境采集信息至监管端，以使监管端基于三维点云重建结果和环境采集信息判断第一规划路径是否符合第三预设指标，以获得判断结果。第三预设指标用于表征对应第一规划路径的三维点云重建结果和环境采集信息是否一致，监管端生成对应的判断结果后，发送判断结果至云端平台。
云端平台接收来自监管端的判断结果，判断结果可以是表征对应第一规划路径的三维点云重建结果和环境采集信息一致的判断结果，也可以是表征对应第一规划路径的三维点云重建结果和环境采集信息不一致的判断结果。当判断结果为对应第一规划路径的三维点云重建结果和环境采集信息一致时，云端平台发送规划反馈信息至机器人，以指示机器人可以将第一规划路径确定为第二规划路径，并根据第二规划路径执行指定操作，在该情况下，第一规划路径与第二规划路径相同。
当判断结果为对应第一规划路径的三维点云重建结果和环境采集信息不一致时，监管端的判断结果中还携带有基于点云生成的路径调整信息，云端平台将基于点云生成的路径调整信息转换为基于地图生成的路径调整信息，并发送规划反馈信息至机器人。该情况下，规划反馈信息中携带有基于地图生成的路径调整信息，以指示机器人基于地图生成的路径调整信息对第一规划路径进行，获得第二规划路径，并根据第二规划路径执行指定操作，在该情况下，第一规划路径与第二规划路径不同。路径调整信息可以是对路径的调整，也可以是障碍物的调整，在本方法中，路径调整信息为对障碍物的调整。
图4为本公开实施例一种基于点云的地图校准方法的场景实现流程示意图；图5为本公开实施例一种基于点云的地图校准方法的场景示意图； 图6为本公开实施例一种基于点云的地图校准方法执行后的场景效果图。
参见图4、图5和图6，为方便上述实施方式的整体性理解，以下提供一种具体实施场景，该具体实施场景包含机器人和云端平台。该场景中，地图校准方法包括：
步骤401，机器人接收到来自用户的规划指令，规划指令用于指示机器人抓取放置在桌面上的水杯；
步骤402，机器人顺着水杯所指向的方向，通过雷达和RGBD摄像头捕获环境采集信息上传至云端机器人控制平台(HARI)的计算机视觉模块；
步骤403，云端计算机视觉模块将机器人上传的环境采集信息进行实时三维点云重建，获得实时三维点云重建结果；
步骤404，云端计算机视觉模块对重建出的实时三维点云重建结果进行识别，获得识别结果，识别结果包括三维点云重建结果中的实时障碍物与二维和三维包围盒，并标记该实时障碍物的类型和置信度。
步骤405，云端计算机视觉模块将实时三维点云重建结果与所有识别结果发送至远程监管端。
步骤406，云端计算机视觉模块通过云端平台将实时三维点云重建结果与识别结果发送至HARI。
步骤407，HARI对计算机视觉模块识别到障碍物进行置信度判断，若障碍物对应的置信度高于0.9(值域为0-1)，则云端机器人控制平台将该障碍物标记为现实环境存在的障碍物，发送与该障碍物对应的地图校准信息至机器人，使机器人在机器人携带的三维地图中对该障碍物进行校准。
步骤408，当障碍物对应的置信度不高于0.9的情况下，则HARI发送障碍确认请求至远程监管端，远程监管端由人工操作员控制，远程监管端接收到障碍确认请求后，会将该障碍物对应的三维点云重建结果渲染至三维地图内，以确定该障碍物是否为现实环境存在的障碍物。
步骤409，远程监管端将确定该障碍物是否为现实环境存在的障碍物的结果发送至HARI，HARI根据该结果确定是否需要发送对应的地图校准信息至机器人，使机器人在机器人携带的三维地图中对该障碍物进行校准。
步骤410，机器人完成三维地图的校准后，根据规划指令进行路径规划，获得第一规划路径，并发送第一规划路径至HARI。
步骤411，HARI根据第一规划路径转换为三维点云重建结果和采集信息后，发送三维点云重建结果和采集信息至远程监管端。
步骤412，远程监管端将该障碍物对应的三维点云重建结果渲染至三维地图内，并根据识别结果中障碍物与二维和三维包围盒大小生成对应类型和大小的3d物体模型，放置在三维地图对应的位置上。
步骤413，通过比对渲染至三维地图内的三维点云重建结果和3d物体模型的匹配度，可以确定现实环境内机器人可抓取物体的类型与大小，并可以根据三维点重建结果和3d物体模型的比较匹配程度，人工操作员可以选择进行物体调整校准与标记，物体的调整校准包括但不限于类型、大小、位置、旋转。例如，发现杯子1的没有被识别到，杯子2的识别结果大小与真实不符，导致在抓取杯子2时，机器人的手可能会与杯子1发生碰撞，从而引发安全风险。故人工操作员在虚拟环境中添加杯子1，并在虚拟环境中调整杯子2的大小，通过点云使虚拟和现实环境校准保持一致。
步骤414，远程监端将调整校准信息发送至HARI，HARI根据调整校准信息转换为基于电子地图的路径调整信息，并发送路径调整信息至机器人，机器人端根据收到的路径调整信息重新规划路径，获得第二规划路径，以指导后续的导航，机器人抓取杯子2时会考虑有杯子1的避障运动规划。
步骤415，机器人执行第二规划路径，完成与规划指令对应的指定该操作，即机器人安全地执行抓取行为。
图7为本公开实施例一种基于点云的地图校准系统的实现模块示意图。
参见图7，本公开实施例另一方面提供一种云端平台，云端平台与指定机器人通信连接，包括：第一获得模块701，用于获得来自指定机器人的环境采集信息；重建模块702，用于对环境采集信息进行三维点云重建，对三维点云重建结果进行障碍识别，获得障碍识别结果；障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息；第一发送模块703，用于当确定为置信度信息满足第一预设指标时，将与障碍物信息对应的地图校准信息发送至指定机器人，以使指定机器人根据地图校准信息对地图信息进行校准。
在一可实施方式中，云端平台还通信连接有监管端，还包括：第一发送模块703，还用于当确定为置信度信息不满足第一预设指标时，发送障碍确认请求至监管端，以请求监管端确定与置信度信息对应的障碍物信息是否满足第二预设指标；第一接收模块704，用于接收来自监管端的障碍反馈信息，障碍反馈信息携带有与障碍物信息对应的确定结果；第一发送模块703，还用于当确定结果为与置信度信息对应的障碍物信息满足第二预设指标时，发送与障碍物信息对应的地图校准信息至指定机器人，以使指定机器人根据地图校准信息对地图信息进行校准。
在一可实施方式中，还包括：第一接收模块704，还用于接收来自指定机器人的指令规划信息，指令规划信息中携带有第一规划路径；第一发送模块703，还用于发送与第一规划路径对应的三维点云重建结果和环境采集信息至监管端，以使监管端基于三维点云重建结果和环境采集信息判断第一规划路径是否符合第三预设指标，以获得判断结果；第一接收模块704，还用于接收来自监管端的判断结果，基于判断结果发送规划反馈信息至指定机器人，以使指定机器人基于规划反馈信息确定第二规划路径；其中，第一规划路径与第二规划路径相同或不同。
本公开实施例另一方面提供一种机器人，机器人与云端平台通信连接， 包括：第二发送模块705，用于发送环境采集信息至云端平台，以使云端平台根据环境采集信息进行三维点云重建，对三维点云重建结果进行障碍识别，获得障碍识别结果；障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息；第二接收模块706，用于接收来自云端平台的地图校准信息，地图校准信息与置信度信息满足第一预设指标的障碍物信息对应；校准模块707，用于根据地图校准信息对地图信息进行校准,获得校准地图信息。
在一可实施方式中，还包括：第二获得模块708，用于获得规划指令，规划指令用于指示根据指定操作在地图信息上进行路径规划；采集模块709，用于基于规划指令进行信息采集，获得环境采集信息。
在一可实施方式中，还包括：规划模块710，用于在校准后的地图信息上对指定操作进行路径规划，获得第一规划路径；第二发送模块705，还用于发送第一规划路径至云端平台，以指示云端平台通过监管端判断第一规划路径是否符合第三预设指标，以获得判断结果；第二接收模块706，还用于接收来自云端平台的规划反馈信息，规划反馈信息与判断结果对应，基于规划反馈信息确定第二规划路径；其中，第一规划路径与第二规划路径相同或不同。
本公开实施例另一方面提供一种基于点云的地图校准系统，系统包括云端平台和指定机器人，云端平台与指定机器人通信连接，云端平台包括：第一获得模块，用于获得来自指定机器人的环境采集信息；重建模块，用于对环境采集信息进行三维点云重建，对三维点云重建结果进行障碍识别，获得障碍识别结果；障碍识别结果包括障碍物信息和与障碍物信息对应的置信度信息；第一发送模块，用于当确定为置信度信息满足第一预设指标时，将与障碍物信息对应的地图校准信息发送至指定机器人，以使指定机器人根据地图校准信息对地图信息进行校准。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本公开的至少一个实施例或示例中。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或隐含地包括至少一个该特征。在本公开的描述中，“多个”的含义是两个或两个以上，除非另有明确具体的限定。
以上，仅为本公开的具体实施方式，但本公开的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本公开的保护范围之内。因此，本公开的保护范围应以权利要求的保护范围为准。

Claims (10)

  1. 一种基于点云的地图校准方法，其特征在于，所述方法应用于云端平台，所述云端平台与指定机器人通信连接，包括：
    获得来自所述指定机器人的环境采集信息；
    对所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；
    当确定为所述置信度信息满足第一预设指标时，将与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
  2. 根据权利要求1所述的方法，其特征在于，所述云端平台还通信连接有监管端，所述方法还包括：
    当确定为所述置信度信息不满足第一预设指标时，发送障碍确认请求至监管端，以请求所述监管端确定与所述置信度信息对应的障碍物信息是否满足第二预设指标；
    接收来自所述监管端的障碍反馈信息，所述障碍反馈信息携带有与所述障碍物信息对应的确定结果；
    当所述确定结果为与所述置信度信息对应的障碍物信息满足第二预设指标时，发送与所述障碍物信息对应的地图校准信息至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
  3. 根据权利要求1所述的方法，其特征在于，在将与所述障碍物信息对应的地图校准信息发送至所述指定机器人之后，所述方法还包括：
    接收来自所述指定机器人的指令规划信息，所述指令规划信息中携带有第一规划路径；
    发送与所述第一规划路径对应的三维点云重建结果和环境采集信息至 监管端，以使监管端基于所述三维点云重建结果和环境采集信息判断所述第一规划路径是否符合第三预设指标，以获得判断结果；
    接收来自所述监管端的判断结果，基于所述判断结果发送规划反馈信息至所述指定机器人，以使所述指定机器人基于所述规划反馈信息确定第二规划路径；
    其中，所述第一规划路径与所述第二规划路径相同或不同。
  4. 一种基于点云的地图校准方法，其特征在于，所述方法应用于机器人，所述机器人与云端平台通信连接，所述方法包括：
    发送环境采集信息至所述云端平台，以使所述云端平台根据所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；
    接收来自所述云端平台的地图校准信息，所述地图校准信息与所述置信度信息满足第一预设指标的障碍物信息对应；
    根据所述地图校准信息对地图信息进行校准。
  5. 根据权利要求4所述的方法，其特征在于，在发送环境采集信息至所述云端平台之前，所述方法还包括：
    获得规划指令，所述规划指令用于指示根据指定操作在所述地图信息上进行路径规划；
    基于所述规划指令进行信息采集，获得环境采集信息。
  6. 根据权利要求5所述的方法，其特征在于，在根据所述地图校准信息对地图信息进行校准之后，所述方法还包括：
    在校准后的地图信息上对所述指定操作进行路径规划，获得第一规划路径；
    发送所述第一规划路径至云端平台，以指示云端平台通过监管端判断 所述第一规划路径是否符合第三预设指标，以获得判断结果；
    接收来自云端平台的规划反馈信息，所述规划反馈信息与所述判断结果对应，基于所述规划反馈信息确定第二规划路径；
    其中，所述第一规划路径与所述第二规划路径相同或不同。
  7. 一种云端平台，其特征在于，所述云端平台与指定机器人通信连接，包括：
    获得模块，用于获得来自指定机器人的环境采集信息；
    重建模块，用于对所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；
    第一发送模块，用于当确定为所述置信度信息满足第一预设指标时，将与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
  8. 根据权利要求7所述的云端平台，其特征在于，所述云端平台还通信连接有监管端，还包括：
    所述第一发送模块，还用于当确定为所述置信度信息不满足第一预设指标时，发送障碍确认请求至监管端，以请求所述监管端确定与所述置信度信息对应的障碍物信息是否满足第二预设指标；
    第一接收模块，用于接收来自所述监管端的障碍反馈信息，所述障碍反馈信息携带有与所述障碍物信息对应的确定结果；
    所述第一发送模块，还用于当所述确定结果为与所述置信度信息对应的障碍物信息满足第二预设指标时，发送与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
  9. 一种机器人，其特征在于，所述机器人与云端平台通信连接，包括：
    第二发送模块，用于发送环境采集信息至所述云端平台，以使所述云端平台根据所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；
    第二接收模块，用于接收来自所述云端平台的地图校准信息，所述地图校准信息与所述置信度信息满足第一预设指标的障碍物信息对应；
    校准模块，用于根据所述地图校准信息对地图信息进行校准,获得校准地图信息。
  10. 一种基于点云的地图校准系统，其特征在于，所述系统包括云端平台和指定机器人，所述云端平台与指定机器人通信连接，所述云端平台包括：
    获得模块，用于获得来自指定机器人的环境采集信息；
    重建模块，用于对所述环境采集信息进行三维点云重建，对所述三维点云重建结果进行障碍识别，获得障碍识别结果；所述障碍识别结果包括障碍物信息和与所述障碍物信息对应的置信度信息；
    第一发送模块，用于当确定为所述置信度信息满足第一预设指标时，将与所述障碍物信息对应的地图校准信息发送至所述指定机器人，以使所述指定机器人根据所述地图校准信息对地图信息进行校准。
PCT/CN2021/122363 2020-11-09 2021-09-30 基于点云的地图校准方法、系统、机器人及云端平台 WO2022095654A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2021578196A JP7465290B2 (ja) 2020-11-09 2021-09-30 点群に基づく地図較正方法、システム、ロボット及びクラウドプラットフォーム
US17/563,792 US20220147049A1 (en) 2020-11-09 2021-12-28 Point cloud-based map calibration method and system, robot and cloud platform

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011237532.9 2020-11-09
CN202011237532.9A CN112327851B (zh) 2020-11-09 2020-11-09 基于点云的地图校准方法、系统、机器人及云端平台

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/563,792 Continuation US20220147049A1 (en) 2020-11-09 2021-12-28 Point cloud-based map calibration method and system, robot and cloud platform

Publications (1)

Publication Number Publication Date
WO2022095654A1 true WO2022095654A1 (zh) 2022-05-12

Family

ID=74316516

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/122363 WO2022095654A1 (zh) 2020-11-09 2021-09-30 基于点云的地图校准方法、系统、机器人及云端平台

Country Status (2)

Country Link
CN (1) CN112327851B (zh)
WO (1) WO2022095654A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112327851B (zh) * 2020-11-09 2023-08-22 达闼机器人股份有限公司 基于点云的地图校准方法、系统、机器人及云端平台
CN112859859B (zh) * 2021-01-13 2022-04-22 中南大学 一种基于三维障碍物体素对象映射的动态栅格地图更新方法
US11624831B2 (en) 2021-06-09 2023-04-11 Suteng Innovation Technology Co., Ltd. Obstacle detection method and apparatus and storage medium
CN113253299B (zh) * 2021-06-09 2022-02-01 深圳市速腾聚创科技有限公司 障碍物检测方法、装置及存储介质
CN114652217B (zh) * 2022-03-02 2023-10-27 美智纵横科技有限责任公司 控制方法、清洁机器人和存储介质
CN115033586B (zh) * 2022-08-11 2022-11-04 上海伯镭智能科技有限公司 一种矿场路边界的感知及众包地图更新方法
CN115665213B (zh) * 2022-12-28 2023-03-10 佛山电力设计院有限公司 一种新设备挂网试运行基地数字孪生系统

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105563484A (zh) * 2015-12-08 2016-05-11 深圳前海达闼云端智能科技有限公司 一种云机器人系统、机器人和机器人云平台
WO2017177444A1 (zh) * 2016-04-15 2017-10-19 深圳前海达闼云端智能科技有限公司 一种云机器人系统、机器人及机器人云平台
CN108387917A (zh) * 2018-01-16 2018-08-10 达闼科技（北京）有限公司 导盲方法、电子设备和计算机程序产品
CN109074668A (zh) * 2018-08-02 2018-12-21 深圳前海达闼云端智能科技有限公司 路径导航方法、相关装置及计算机可读存储介质
US20190035096A1 (en) * 2017-07-25 2019-01-31 Shenzhen University Method and apparatus of scene reconstruction
US20190187703A1 (en) * 2017-12-19 2019-06-20 X Development Llc Semantic Obstacle Recognition For Path Planning
CN111367289A (zh) * 2020-03-19 2020-07-03 上海陆根智能传感技术有限公司 一种人工智能设备
CN111899299A (zh) * 2020-06-16 2020-11-06 弗徕威智能机器人科技（上海）有限公司 地面障碍物地图标记方法、移动机器人和存储介质
CN112327851A (zh) * 2020-11-09 2021-02-05 达闼机器人有限公司 基于点云的地图校准方法、系统、机器人及云端平台

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108763571B (zh) * 2018-06-05 2021-02-05 北京智行者科技有限公司 一种作业地图更新方法
CN111368607B (zh) * 2018-12-26 2023-06-02 北京欣奕华科技有限公司 一种机器人、障碍物的检测方法及检测装置
CN110531759B (zh) * 2019-08-02 2020-09-22 深圳大学 机器人探索路径生成方法、装置、计算机设备和存储介质
CN111766862B (zh) * 2019-10-28 2022-12-27 广州极飞科技股份有限公司 避障控制方法、装置、电子设备和计算机可读存储介质
CN111123949B (zh) * 2019-12-31 2023-07-07 达闼机器人股份有限公司 机器人的避障方法、装置、机器人及存储介质
CN111426326B (zh) * 2020-01-17 2022-03-08 深圳市镭神智能系统有限公司 一种导航方法、装置、设备、系统及存储介质
CN111461245B (zh) * 2020-04-09 2022-11-04 武汉大学 一种融合点云和图像的轮式机器人语义建图方法及系统

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105563484A (zh) * 2015-12-08 2016-05-11 深圳前海达闼云端智能科技有限公司 一种云机器人系统、机器人和机器人云平台
WO2017177444A1 (zh) * 2016-04-15 2017-10-19 深圳前海达闼云端智能科技有限公司 一种云机器人系统、机器人及机器人云平台
US20190035096A1 (en) * 2017-07-25 2019-01-31 Shenzhen University Method and apparatus of scene reconstruction
US20190187703A1 (en) * 2017-12-19 2019-06-20 X Development Llc Semantic Obstacle Recognition For Path Planning
CN108387917A (zh) * 2018-01-16 2018-08-10 达闼科技（北京）有限公司 导盲方法、电子设备和计算机程序产品
CN109074668A (zh) * 2018-08-02 2018-12-21 深圳前海达闼云端智能科技有限公司 路径导航方法、相关装置及计算机可读存储介质
CN111367289A (zh) * 2020-03-19 2020-07-03 上海陆根智能传感技术有限公司 一种人工智能设备
CN111899299A (zh) * 2020-06-16 2020-11-06 弗徕威智能机器人科技（上海）有限公司 地面障碍物地图标记方法、移动机器人和存储介质
CN112327851A (zh) * 2020-11-09 2021-02-05 达闼机器人有限公司 基于点云的地图校准方法、系统、机器人及云端平台

Also Published As

Publication number Publication date
CN112327851B (zh) 2023-08-22
CN112327851A (zh) 2021-02-05

Similar Documents

Publication Publication Date Title
WO2022095654A1 (zh) 基于点云的地图校准方法、系统、机器人及云端平台
KR102170748B1 (ko) ＩｏＴ 기반 스마트 건설을 위한 3차원 공간정보 현장 관리 플랫폼 시스템
CN108876822B (zh) 一种行为危险度评估方法及家庭安防看护系统
US8854594B2 (en) System and method for tracking
CN109240311B (zh) 基于智能机器人的室外电力场地施工作业监督方法
CN111459166A (zh) 一种灾后救援环境下含受困人员位置信息的情景地图构建方法
CN110850723A (zh) 一种基于变电站巡检机器人系统的故障诊断及定位方法
CN109470238A (zh) 一种定位方法、装置和移动终端
WO2021036587A1 (zh) 面向电力巡检场景的定位方法及系统
CN109284928A (zh) 一种基于无人机技术的建筑施工监控方法和系统
CN116507286A (zh) 超声图像数据采集分析识别一体化机器人，平台
CN111489392B (zh) 多人环境下单个目标人体运动姿态捕捉方法及系统
CN109931940B (zh) 一种基于单目视觉的机器人定位置信度评估方法
CN109754419B (zh) 位置推测系统
KR20190063967A (ko) 스테레오 카메라와 3d 바코드를 이용한 위치 측정 방법 및 장치
KR101380852B1 (ko) 사용자로부터 환경에 대한 사진 입력을 받는 이동로봇의 ｓｌａｍ 시스템 및 방법
CN114115277A (zh) 一种基于巡检机器人的巡检管理方法及相关设备
CN107160402A (zh) 一种机械臂移动路径规划系统及方法
US20160239697A1 (en) Manager, management system, management method, and non-transitory computer readable storage medium
JP7465290B2 (ja) 点群に基づく地図較正方法、システム、ロボット及びクラウドプラットフォーム
CN114786611A (zh) 用于减轻机器人系统的碰撞的系统和方法
CN109948192B (zh) 一种吊装作业空间建模方法及系统
CN116429112A (zh) 多机器人协同定位方法和装置、设备及存储介质
CN109754665A (zh) 一种机械制造模具虚拟装配培训信息处理系统及培训方法
JPWO2019244434A1 (ja) 情報処理装置、情報処理方法及びプログラム

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021578196

Country of ref document: JP

Kind code of ref document: A

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

Ref document number: 21888351

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: 21888351

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 1205A DATED 04/10/2024)