WO2022120670A1 - 机械臂的运动轨迹规划方法及装置、机械臂及存储介质 - Google Patents

机械臂的运动轨迹规划方法及装置、机械臂及存储介质 Download PDF

Info

Publication number
WO2022120670A1
WO2022120670A1 PCT/CN2020/135084 CN2020135084W WO2022120670A1 WO 2022120670 A1 WO2022120670 A1 WO 2022120670A1 CN 2020135084 W CN2020135084 W CN 2020135084W WO 2022120670 A1 WO2022120670 A1 WO 2022120670A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
initial
motion trajectory
motion
target
Prior art date
Application number
PCT/CN2020/135084
Other languages
English (en)
French (fr)
Inventor
郑大可
刘益彰
庞建新
谭欢
熊友军
Original Assignee
深圳市优必选科技股份有限公司
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 深圳市优必选科技股份有限公司 filed Critical 深圳市优必选科技股份有限公司
Priority to PCT/CN2020/135084 priority Critical patent/WO2022120670A1/zh
Priority to US17/566,726 priority patent/US20220184808A1/en
Publication of WO2022120670A1 publication Critical patent/WO2022120670A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
    • B25J9/1697—Vision controlled systems
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1602—Programme controls characterised by the control system, structure, architecture
    • B25J9/1605—Simulation of manipulator lay-out, design, modelling of manipulator
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1661—Programme controls characterised by programming, planning systems for manipulators characterised by task planning, object-oriented languages
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
    • B25J9/1666—Avoiding collision or forbidden zones
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/40—Robotics, robotics mapping to robotics vision
    • G05B2219/40102—Tasks are classified in types of unit motions
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/40—Robotics, robotics mapping to robotics vision
    • G05B2219/40116—Learn by operator observation, symbiosis, show, watch
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/40—Robotics, robotics mapping to robotics vision
    • G05B2219/40395—Compose movement with primitive movement segments from database
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/40—Robotics, robotics mapping to robotics vision
    • G05B2219/40477—Plan path independent from obstacles, then correction for obstacles

Definitions

  • the invention relates to the technical field of intelligent control, and in particular, to a motion trajectory planning method and device of a robotic arm, a robotic arm and a storage medium.
  • the main purpose of the present invention is to provide a motion trajectory planning method, device, device and storage medium for a robotic arm, which can solve the problem that the robotic arm in the prior art cannot plan the motion trajectory autonomously and flexibly.
  • a first aspect of the present invention provides a method for planning a motion trajectory of a robotic arm, the method comprising:
  • the robotic arm receives the task instruction, obtain the environmental data collected by the vision detection system of the robotic arm;
  • the environment data and the preset teaching motion dynamic system DS model library, the initial DS model motion trajectory of the robotic arm is determined, and the teaching motion DS model library at least includes human-based teaching activities The generated DS model motion trajectory;
  • the movement trajectory of the initial DS model is corrected to obtain the target movement trajectory of the robotic arm.
  • a motion trajectory planning device for a robotic arm comprising:
  • an acquisition module configured to acquire the environmental data collected by the visual detection system of the robotic arm if the robotic arm receives the task instruction
  • a determination module configured to determine the initial DS model motion trajectory of the robotic arm according to the task instruction, the environmental data and the preset teaching motion dynamic system DS model library, and the teaching motion dynamic system DS model library At least include the DS model motion trajectory generated based on human teaching activities;
  • the correction module is used for correcting the motion trajectory of the initial DS model to obtain the target motion trajectory of the robotic arm.
  • a third aspect of the present invention provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, causes the processor to execute the method in the first aspect. each step.
  • a fourth aspect of the present invention provides a robotic arm, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes as described in Section 1. The various steps in the method of one aspect.
  • the present invention provides a motion trajectory planning method for a robotic arm. After the robotic arm receives a task instruction, it acquires environmental data collected by a visual detection system of the robotic arm, and according to the above-mentioned task instruction, environmental data and preset teaching motion Dynamic system (Dynamic System, DS) model library, determine the initial DS model motion trajectory of the robotic arm, and revise the initial DS model motion trajectory to obtain the target motion trajectory of the robotic arm, the target motion trajectory is matching the described The trajectory of the task command.
  • DS teaching motion Dynamic system
  • the robot arm can use the teaching motion DS model library to determine the initial DS model motion trajectory, and further analyze the motion trajectory of the DS model.
  • the motion trajectory of the initial DS model is corrected to obtain the target motion trajectory that meets the requirements of the task instruction, and realizes the planning of the autonomous and flexible motion trajectory of the robotic arm.
  • FIG. 1 is a schematic flowchart of a method for planning a motion trajectory of a robotic arm according to an embodiment of the present invention
  • FIG. 2 is another flowchart of a method for planning a motion trajectory of a robotic arm according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of the refinement step of step 103 in the embodiment shown in FIG. 1 of the present invention
  • FIG. 4 is a schematic diagram of a concave obstacle in an embodiment of the present invention.
  • Fig. 5 is another schematic diagram of the concave obstacle shown in Fig. 4;
  • FIG. 6 is a schematic diagram of a system block diagram of a robotic arm in an embodiment of the present invention.
  • FIG. 1 is a method for planning a motion trajectory of a robotic arm in an embodiment of the present invention.
  • the method includes:
  • Step 101 If the robotic arm receives the task instruction, obtain the environmental data collected by the vision detection system of the robotic arm;
  • the above-mentioned motion trajectory planning method of the robotic arm is implemented by a motion trajectory planning device of the mechanical arm, and the motion trajectory planning device is a program module, and the program module is stored in the computer-readable storage medium of the mechanical arm,
  • the processor in the robotic arm can call and execute the program module to implement the above-mentioned method for planning the motion trajectory of the robotic arm.
  • the user or the device can send a task command to the robotic arm, and the task command can be a voice command or a non-voice command.
  • the user can say the command by voice, and the voice collection module of the robotic arm can interpret the user's voice. Collect and generate task instructions, or the robotic arm can receive task instructions sent by other users or devices through the network.
  • the robotic arm has a visual detection system, which can detect environmental data in the movable space of the robotic arm.
  • the visual detection system can include a camera, and the above-mentioned environmental data is obtained by capturing images with the camera. For example, if there is a water cup in the movable space of the robotic arm, an image containing the water cup can be captured by the camera included in the visual inspection system.
  • Step 102 according to the task instruction, the environment data and the preset teaching motion DS model library, determine the initial DS model motion trajectory of the robotic arm, and the teaching motion DS model library at least includes the DS model motion trajectory generated based on the human teaching activity;
  • Step 103 correcting the motion trajectory of the initial DS model to obtain the target motion trajectory of the robotic arm.
  • the motion trajectory of the manipulator refers to the movement trajectory of the end of the manipulator, and the motion trajectory of the manipulator can be described based on the DS principle, wherein DS is a mathematical concept, and in the DS system there is a
  • DS is a mathematical concept
  • the fixed rule describes the time evolution of a point in the geometric space. Therefore, the motion trajectory of the end of the manipulator (the end is regarded as a point) can be described by the DS principle.
  • a teaching motion DS model library is preset.
  • the above-mentioned teaching motion DS model library contains at least the motion trajectory of the DS model generated based on the human teaching activity, wherein, the human teaching activity refers to the actual completion of a task by a human using an arm, and the task can be composed of an action, or can be Consists of multiple actions, for example, the human teaching activity can be: raising the arm, lowering the arm, picking up a cup, opening the refrigerator door, etc., and can be actually demonstrated by a person performing the teaching activity "raise the arm” , "put down the arm”, “pick up the refrigerator” and “open the refrigerator door”, and capture the video data of the person demonstrating the above actions, and confirm the position of the end of the arm in the video data based on the DS principle, The movement trajectory of the end of the arm in the process of performing the above actions is obtained.
  • the arm of the above-mentioned person corresponds to the mechanical arm, and the palm part of the arm is the end of the arm, which corresponds to the end of the mechanical arm.
  • the mechanical arm can simulate the movement of the above-mentioned personnel.
  • the movement trajectory of the end of the arm obtained based on the DS principle can be used as the movement trajectory of the DS model of the robot arm.
  • the movement trajectory of the DS model corresponding to the corresponding action can be obtained based on the movement trajectory, so that the movement trajectory of the DS model corresponding to the corresponding action can be obtained through
  • the above-mentioned teaching motion DS model library is obtained, and the robotic arm can perform tasks by simulating the real actions of human beings, so as to have the ability to determine the motion trajectory autonomously and flexibly, and improve the autonomy and flexibility of the robotic arm .
  • teaching motion DS model libraries can be set in different scenarios.
  • a human in a home scene, a human can teach the possible actions of activities at home, and the home scene can be obtained based on the actions taught by humans.
  • a corresponding teaching motion DS model library in practical applications, a corresponding teaching motion DS model library can be set according to specific scenarios and requirements, which is not limited here.
  • the initial DS model motion trajectory of the manipulator can be determined according to the task instruction, environmental data and the above-mentioned preset teaching motion DS model library, and considering that the teaching motion DS model library is actually equivalent to a
  • the template library is also different from the real activities of the manipulator. For example, there may be obstacles, and obstacle avoidance may be required, or the end positions of the tasks may be different, etc. Therefore, after obtaining the above initial DS model motion trajectory After that, the motion trajectory of the initial DS model will be corrected to obtain the target motion trajectory of the robotic arm.
  • the robot arm can be tracked and controlled, so that the robot arm can accurately track the target movement trajectory, so that the robot arm can realize the tasks indicated by the above task instructions and realize the mechanical Human-like autonomous movement of the arm.
  • the robotic arm by generating the motion trajectory of the DS model based on human teaching activities, and forming the DS model library of the teaching motion dynamic system, the robotic arm can use the teaching motion DS model library to determine the initial DS model motion trajectory , and further revise the motion trajectory of the initial DS model to obtain the target motion trajectory that meets the requirements of the task command, and realize the autonomous and flexible motion trajectory planning of the robotic arm.
  • FIG. 2 is another schematic flowchart of a method for planning a motion trajectory of a robotic arm in an embodiment of the present invention, including:
  • Step 201 if the robotic arm receives the task instruction, then obtain the environmental data collected by the visual inspection system of the robotic arm;
  • step 201 is similar to the content described in step 101 in the embodiment shown in FIG. 1 .
  • step 101 is similar to the content described in step 101 in the embodiment shown in FIG. 1 .
  • step 101 is similar to the content described in step 101 in the embodiment shown in FIG. 1 .
  • step 101 is similar to the content described in step 101 in the embodiment shown in FIG. 1 .
  • step 101 is similar to the content described in step 101 in the embodiment shown in FIG. 1 .
  • Step 202 parse the task instruction, determine the target task name and the target task object indicated by the task instruction, and the target task object refers to the operated object pointed to by the robotic arm executing the task instruction;
  • Step 203 Perform object recognition on the environmental data, and determine the first object included in the environmental data;
  • Step 204 Determine the initial DS model motion trajectory of the robotic arm according to the target task name, the target task object, the first object and the teaching motion DS model library;
  • the above-mentioned teaching motion DS model library includes the correspondence between task names, task objects and DS model motion trajectories, wherein, when obtaining the DS model motion trajectory based on human teaching motion, the Set the task object and task name.
  • the task name can be "pick up”
  • the task object can be "water cup”, “bottle”, “scissors", “mobile phone”, etc.
  • the task command can be: “pick up” water glass”, “pick up the bottle”, “pick up the scissors”, “pick up the phone”, etc.
  • the task object can be empty, that is, the task performed by the robotic arm can be the action of the arm, which has nothing to do with external objects.
  • the task name can be "raise hand", “point forward” Etc., in practical applications, whether the task object is empty in the above-mentioned corresponding relationship is related to a specific task, and can be set as required, which is not limited here.
  • the motion trajectory of the DS model in the above-mentioned teaching motion DS model library may specifically be: the video data for task A in the human teaching activity is obtained, and the teaching motion dynamic system model identification method based on the Gaussian mixture model is used for the teaching motion.
  • the video data is identified and processed to obtain the motion trajectory of the DS model corresponding to task A, and it can be understood that the task name and task object of the motion trajectory of the DS model can be set based on the analysis of task A.
  • the task instruction may be parsed to determine the target task name and target task object indicated by the task instruction, where the target task object refers to The manipulated object pointed to by the robot arm executing the task command. For example, if the task command is "pick up the water cup”, the target task name can be "pick up”, and the target task object is "water cup”.
  • the information contained in the task command can be obtained, and the target task name and target task object in the information can be determined.
  • a user can input a task on the display interface of a terminal that can communicate with the robotic arm, that is, "open the refrigerator", and the terminal generates a task instruction, and the terminal sends the task instruction to the robotic arm through the network, and the robotic arm responds to the After parsing the task instruction, it can be determined that the target task name is "open” and the target task object is "refrigerator".
  • the above task command is a voice command
  • Target task name and target task object If the above task command is a voice command, it is necessary to first perform voice recognition on the voice command, determine the text content contained in the voice command, and further divide the text content by keywords, determine the keywords, and then determine the above-mentioned keywords based on the keywords.
  • object recognition will be performed on the environmental data to determine the first object contained in the environmental data, wherein the above-mentioned first object refers to all objects existing in the image contained in the environmental data.
  • the initial DS model motion trajectory of the robotic arm will be determined according to the above-mentioned target task name, target task object, the first object and the teaching motion DS model library, wherein, due to the teaching motion DS model
  • the library contains task names, task objects and DS model motion trajectories. Therefore, the above-mentioned initial DS model motion trajectories can be obtained as follows:
  • Step a judging whether the first object contains the target task object
  • Step b When the first object includes the target task object, use the target task name and the target task object to search the teaching motion DS model library, and determine the initial DS model motion trajectory corresponding to the target task name and the target task object.
  • the first object contains the target task object. Specifically, if the target task object is not empty, the first object is traversed, and after the traversal is completed, the second object in the traversed objects that is the same as the target task object is determined. , then it is determined that the second object is the target task object. For example, if the first object includes a water cup, scissors, a pen, a notebook, etc., and the target task object is a "water cup", then traverse the first object to determine that it contains the target task object "water cup” ". It can be understood that when the target task object is empty, it can also be determined that the first object contains the target task object.
  • teaching motion DS model library will be searched by using the target task name and the target task object, and the initial DS model motion trajectory corresponding to the target task name and the target task object will be determined.
  • the task indicated by the task instruction may be composed of multiple small tasks.
  • the task is "open the refrigerator"
  • the task can be decomposed into two steps for execution, one is to move to the refrigerator door,
  • the corresponding target task name is: moving position
  • the target task object is: refrigerator
  • the other is to perform the opening action
  • the corresponding target task name is: open
  • the target task object is refrigerator, in this case, then It is necessary to obtain the motion trajectories of the initial DS model corresponding to the two steps respectively, and use the initial DS model motion trajectories of the two steps as the motion trajectories of the tasks indicated by the task instructions.
  • the division method of tasks and the task size corresponding to the motion trajectory of each DS model in the teaching motion DS model library are not limited here.
  • Step 205 Correct the motion trajectory of the initial DS model to obtain the target motion trajectory of the robotic arm.
  • the initial DS model motion trajectory corresponding to the target task name and the target task object can be effectively determined.
  • FIG. 3 is a schematic flowchart of the refinement step of step 103 in the embodiment shown in FIG. 1 according to an embodiment of the present invention.
  • the refinement step includes:
  • Step 301 judge whether there is an obstacle whose pose is located on the motion trajectory of the initial DS model in the first object included in the environmental data, and obtain a first judgment result; and judge whether the first pose of the target task object in the task instruction is the same as the initial position.
  • the second pose of the target object in the motion trajectory of the DS model is the same, and the second judgment result is obtained;
  • Step 302 revise the motion trajectory of the initial DS model according to the first judgment result and the second judgment result to obtain the target motion trajectory.
  • the initial DS model motion trajectory after the initial DS model motion trajectory is obtained, it will be corrected, so that the target motion trajectory obtained after the correction can be more in line with the trajectory required by the robotic arm to execute the task instruction in the actual scene.
  • the first object included in the above environmental data is determined by performing object recognition on the environmental data. In addition to determining whether there is a target task object, it is also used to determine whether there is an obstacle. It can be understood that the mechanical The arm can determine the pose of the first object included in the environment data, and the pose and the motion trajectory of the DS model are both determined based on the same Cartesian coordinate system.
  • the method of determining the obstacle may be to judge whether there is an obstacle whose pose is located on the motion trajectory of the initial DS model in the first object, and obtain a first judgment result, and the first judgment result may be that it does not exist, or it may be that it exists. , and it is object A. In this way, it can be determined whether there is an obstacle. It can be understood that the determination of the obstacle can be real-time, that is, the position that the end is about to reach on the moving trajectory of the end of the robot arm is determined in real time. whether there are obstacles.
  • the target object in the motion trajectory of the initial DS model refers to the object used in the human teaching activity. For example, if the human teaching activity is "pick up the water glass", the target object is the water glass.
  • the initial DS model motion trajectory is corrected for obstacle avoidance to obtain the target motion trajectory .
  • the initial DS model motion trajectory is corrected for obstacle avoidance, and based on the first pose
  • the initial DS model motion trajectory after obstacle avoidance correction is used to correct the pose of the target object to obtain the target motion trajectory;
  • the initial DS model motion trajectory is determined as the target motion trajectory
  • the above-mentioned obstacle avoidance correction on the motion trajectory of the initial DS model includes: first, based on the shape of the obstacle, determining whether the obstacle is a concave obstacle or a convex obstacle, wherein the concave obstacle
  • An object refers to an object whose shape can be divided into at least two parts by the tangent of the outer edge of the object's shape. Obstacles other than concave obstacles can be called convex obstacles.
  • the concave obstacle when the obstacle is a concave obstacle, the concave obstacle may be divided into at least two convex obstacles that intersect in pairs, wherein the intersecting position of the two intersecting convex obstacles is an intersection line.
  • the combined modal matrix formed by the combination of convex obstacles is calculated, and the combined modal matrix is used to correct the initial DS model motion trajectory.
  • a normal vector and a second normal vector construct a modal matrix corresponding to the intersection point, and use the modal matrix corresponding to the intersection point to correct the motion trajectory of the initial DS model.
  • FIG. 4 is a schematic diagram of a concave obstacle in the embodiment of the present invention.
  • the concave obstacle is an electric drill as an example, which consists of three convex-shaped obstacles.
  • the three convex obstacles are respectively convex obstacle 1, convex obstacle 2 and convex obstacle 3, wherein the convex obstacle 1 and the convex obstacle 2 intersect each other two by two.
  • the convex obstacle 2 and the convex obstacle 3 intersect each other, and the intersecting position of the two intersecting convex obstacles is the intersection line.
  • intersection lines are A and B, where the intersection line The range of A is from point c1 to point c2, and the range of intersection line B is from point c3 to point c4. It can be understood that, for any concave obstacle, it can be divided into a plurality of convex obstacles.
  • the combined modal matrix of the three convex obstacles divided by the above electric drill can be calculated according to the following formula, as follows:
  • the above-mentioned combined modal matrix can be used to correct the initial DS model motion trajectory.
  • FIG. 5 is another schematic diagram of the concave obstacle shown in FIG. 4 , wherein, represents the center point of convex obstacle 1, represents the center point of convex obstacle 2, represents the center point of convex obstacle 3, represents the intersection, and the intersection is located on the surface of the third convex obstacle, represents the normal vector of the third convex obstacle, and Represents the base vector of the hyperplane corresponding to the normal vector of the third convex obstacle.
  • the initial DS model motion trajectory When the initial DS model motion trajectory reaches the intersection of the concave obstacle and is on the intersection line, then determine the first normal vector of the first convex obstacle and the second normal vector of the second convex obstacle that form the intersection line, using The first normal vector and the second normal vector construct a modal matrix corresponding to the intersection point, and the initial DS model motion trajectory is corrected by using the modal matrix corresponding to the intersection point, so that concave obstacles can be avoided.
  • the point of intersection is the point on the intersecting line For example, and point On the intersection of the mth convex obstacle and the nth convex obstacle, then the point
  • the first normal vector at the mth convex obstacle is as follows:
  • b represents the mark of a point on the intersection of the mth convex obstacle and the nth convex obstacle.
  • first normal vector and second normal vector After obtaining the above-mentioned first normal vector and second normal vector, use the first normal vector and the second normal vector to construct an intersection point corresponding The modal matrix, as follows:
  • pin represents the pseudo-inverse.
  • f( ⁇ ) represents the initial movement trajectory
  • Representation based on intersection point Determine the target modal matrix
  • FIG. 5 is another schematic diagram of the concave obstacle shown in FIG. 4 , wherein, represents the center point of convex obstacle 1, represents the center point of convex obstacle 2, represents the center point of convex obstacle 3,
  • the point on the intersection line (intersection line) of the first convex obstacle 1 and the second convex obstacle 2 is shown, and this point is the above-mentioned intersection point.
  • Representation point the normal vector on the first convex obstacle Representation point
  • the normal vector on the second convex obstacle, e 12 means both perpendicular to and , that is, the above value of .
  • the motion trajectory after the obstacle correction is the target motion trajectory. If the obstacle is corrected, the pose of the target object needs to be corrected. If it is corrected, the pose of the target object is further determined, and the corrected motion trajectory is used as the target motion trajectory.
  • the pose of the target object is corrected by replacing the pose of the target object with the first pose of the target task object.
  • the present invention by using environmental data and task instructions, it can be effectively determined whether it is necessary to perform obstacle avoidance correction and the pose correction of the target task object.
  • the motion trajectory of the initial DS model it can effectively achieve obstacle avoidance, and when the pose of the target object needs to be corrected, the pose of the target object will also be corrected.
  • the motion trajectory is more in line with the task requirements in the actual scene, so that the task can be realized effectively and accurately.
  • the technical solutions in the embodiments of the present invention are suitable for scenarios such as elderly care, disability assistance, family services, human-machine collaboration, etc., which can enable the robotic arm to autonomously and flexibly realize the planning of the motion trajectory, and by moving the determined target.
  • the tracking control of the trajectory enables the manipulator to achieve precise control, effectively and accurately complete the desired task, and realize the human-like autonomous motion of the manipulator.
  • the above-mentioned planning process of the motion trajectory is a real-time process, and the real-time planning of the motion trajectory can also be realized.
  • Figure 6 shows a schematic diagram of a system block diagram of a robotic arm in one embodiment.
  • the robotic arm includes a processor, memory and network interface connected through a system bus.
  • the memory includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium of the robotic arm stores an operating system, and also stores a computer program.
  • the processor can implement the motion trajectory planning method of the robotic arm.
  • a computer program may also be stored in the internal memory, and when the computer program is executed by the processor, the processor may execute the method for planning the motion trajectory of the robot arm.
  • FIG. 6 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the robot arm to which the solution of the present application is applied. Include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.
  • a motion trajectory planning device for a robotic arm includes: a processor and a memory, where a computer program is stored in the memory, and when the processor executes the computer program, the processor executes the following steps:
  • the robotic arm receives the task command, obtain the environmental data collected by the vision detection system of the robotic arm;
  • environmental data and the preset teaching motion dynamic system DS model library determine the initial DS model motion trajectory of the manipulator, and the teaching motion DS model library at least contains the DS model motion trajectory generated based on human teaching activities;
  • the movement trajectory of the initial DS model is corrected to obtain the target movement trajectory of the manipulator.
  • a robotic arm comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps:
  • the robotic arm receives the task instruction, obtain the environmental data collected by the vision detection system of the robotic arm;
  • the environment data and the preset teaching motion dynamic system DS model library, the initial DS model motion trajectory of the robotic arm is determined, and the teaching motion DS model library at least includes human-based teaching activities The generated DS model motion trajectory;
  • the movement trajectory of the initial DS model is corrected to obtain the target movement trajectory of the robotic arm.
  • a computer-readable storage medium which stores a computer program, and when the computer program is executed by a processor, causes the processor to perform the following steps:
  • the robotic arm receives the task instruction, obtain the environmental data collected by the vision detection system of the robotic arm;
  • the environment data and the preset teaching motion dynamic system DS model library, the initial DS model motion trajectory of the robotic arm is determined, and the teaching motion DS model library at least includes human-based teaching activities The generated DS model motion trajectory;
  • the movement trajectory of the initial DS model is corrected to obtain the target movement trajectory of the robotic arm.
  • the robotic arm by generating the motion trajectory of the DS model based on human teaching activities, and forming the DS model library of the teaching motion dynamic system, the robotic arm can use the teaching motion DS model library to determine the initial DS model motion trajectory , and further revise the motion trajectory of the initial DS model to obtain the target motion trajectory that meets the requirements of the task command, and realize the autonomous and flexible motion trajectory planning of the robotic arm.
  • Nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Road (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
  • SRAM static RAM
  • DRAM dynamic RAM
  • SDRAM synchronous DRAM
  • DDRSDRAM double data rate SDRAM
  • ESDRAM enhanced SDRAM
  • SLDRAM synchronous chain Road (Synchlink) DRAM
  • SLDRAM synchronous chain Road (Synchlink) DRAM
  • Rambus direct RAM
  • DRAM direct memory bus dynamic RAM
  • RDRAM memory bus dynamic RAM

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Manipulator (AREA)

Abstract

一种机械臂的运动轨迹规划方法及装置、机械臂及存储介质，方法包括：若机械臂接收到任务指令，则获取机械臂的视觉检测系统采集到的环境数据（101），根据任务指令、环境数据及预设的示教运动DS模型库，确定该机械臂的初始DS模型运动轨迹，并对所述初始DS模型运动轨迹进行修正，得到机械臂的目标运动轨迹（103），该目标运动轨迹是匹配所述任务指令的运动轨迹。该方法通过基于人类示教活动生成DS模型运动轨迹，并形成示教运动动态系统DS模型库的方式，使得机械臂能够利用该示教运动DS模型库确定初始DS模型运动轨迹，并进一步地对该初始DS模型运动轨迹进行修正，得到符合任务指令需求的目标运动轨迹，实现机械臂自主且灵活的运动轨迹的规划。

Description

机械臂的运动轨迹规划方法及装置、机械臂及存储介质 技术领域
本发明涉及智能控制技术领域，尤其涉及一种机械臂的运动轨迹规划方法及装置、机械臂及存储介质。
背景技术
随着科技的快速发展，机械臂的技术也越来越成熟，目前，机械臂技术在无人搬运、物流分拣、流水线制造等领域已经取得了很大的提升并得到了成熟的应用，但是在诸如养老、助残、家庭服务、人机协同、复杂未知场景自主操作等领域的提升和应用还有很长的路要走，对于上述的复杂领域，机械臂无法自主且灵活地进行运动轨迹的规划。
发明内容
本发明的主要目的在于提供一种机械臂的运动轨迹规划方法及装置、设备及存储介质，可以解决现有技术中的机械臂无法自主且灵活地进行运动轨迹的规划的问题。
为实现上述目的，本发明第一方面提供一种机械臂的运动轨迹规划方法，所述方法包括：
若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
为实现上述目的，本发明第二方面提供一种机械臂的运动轨迹规划装置，所述装置包括：
获取模块，用于若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
确定模块，用于根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动动态系统DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
修正模块，用于对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
为实现上述目的，本发明第三方面提供一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时，使得所述处理器执行如第一方面所述的方法中的各个步骤。
为实现上述目的，本发明第四方面提供一种机械臂，包括存储器和处理器，所述存储器存储有计算机程序，所述计算机程序被所述处理器执行时，使得所述处理器执行如第一方面所述的方法中的各个步骤。
采用本发明实施例，具有如下有益效果：
本发明提供一种机械臂的运动轨迹规划方法，在机械臂接收到任务指令之后，获取机械臂的视觉检测系统采集到的环境数据，根据上述的任务指令、环境数据及预设的示教运动动态系统(Dynamical System，DS)模型库，确定该机械臂的初始DS模型运动轨迹，并对所述初始DS模型运动轨迹进行修正，得到机械臂的目标运动轨迹，该目标运动轨迹是匹配所述任务指令的运动轨迹。通过基于人类示教活动生成DS模型运动轨迹，并形成示教运动动态系统DS模型库的方式，使得机械臂能够利用该示教运动DS模型库确定初始DS模型运动轨迹，并进一进地对该初始DS模型运动轨迹进行修正，得到符合任务指令需求的目标运动轨迹，实现机械臂自主且灵活的运动轨迹的规划。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施 例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
其中：
图1为本发明实施例中机械臂的运动轨迹规划方法的一流程示意图；
图2为本发明实施例中机械臂的运动轨迹规划方法的另一流程图；
图3为本发明图1所示实施例中步骤103的细化步骤的流程示意图；
图4为本发明实施例中凹形障碍物的示意图；
图5为图4所示凹形障碍物的另一示意图；
图6为本发明实施例中机械臂的系统方框图的示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明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发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
请参阅图1，为本发明实施例中机械臂的运动轨迹规划方法，所述方法包括：
步骤101、若机械臂接收到任务指令，则获取机械臂的视觉检测系统采集到的环境数据；
在本发明实施例中，上述的机械臂的运动轨迹规划方法由机械臂的运动轨迹规划装置实现，该运动轨迹规划装置为程序模块，该程序模块存储在机械臂的计算机可读存储介质中，机械臂中的处理器可以调用并执行该程序模块，以实现上述的机械臂的运动轨迹规划方法。
其中，用户或者设备可以向机械臂发送任务指令，该任务指令可以是语音指令，也可以是非语音指令，例如，用户可以通过语音的方式说出指令，由机 械臂的语音采集模块对用户的语音进行采集，并生成任务指令，或者，机械臂可以接收到其他用户端或者设备通过网络发送的任务指令。
其中，机械臂具有视觉检测系统，该视觉检测系统可以检测到机械臂的可活动空间内的环境数据，该视觉检测系统可以包含摄像头，通过摄像头拍摄图像的方式获取到上述的环境数据。例如，机械臂的可活动空间内存在一个水杯，则利用视觉检测系统包含的摄像头则可以拍摄到包含该水杯的图像。
步骤102、根据任务指令、环境数据及预设的示教运动DS模型库，确定机械臂的初始DS模型运动轨迹，示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
步骤103、对初始DS模型运动轨迹进行修正，得到机械臂的目标运动轨迹。
在本发明实施例中，机械臂的运动轨迹是指机械臂的末端的移动轨迹，可以基于DS原理描述该机械臂的运动轨迹，其中，DS是数学上的一个概念，在DS系统中存在一个固定规则，描述了几何空间中的一个点随时间演化情况，因此，可以通过DS原理描述机械臂的末端(将该末端看作是一个点)的运动轨迹。
在本发明实施例中，预先设置示教运动DS模型库。上述的示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹，其中，人类示教活动是指人类使用手臂真实的完成一个任务，该任务可以由一个动作构成，也可以是由多个动作构成，例如，该人类示教活动可以是：举起手臂，放下手臂，拿起杯子、打开冰箱门等等，可以由一个进行示教活动的人员真实的演示“举起手臂”、“放下手臂”、“拿起冰箱”及“打开冰箱门”的操作，并拍摄得到该人员演示上述动作的视频数据，并基于DS原理对该视频数据中的手臂的末端的位置进行确认，得到手臂的末端在执行上述动作的过程中的移动轨迹，上述人员的手臂则对应着机械臂，手臂的手掌部分则为手臂的末端，且对应着机械臂的末端，机械臂可以模拟上述人员的手臂执行任务，即可将基于DS原理得 到的手臂的末端的移动轨迹作为机械臂的DS模型运动轨迹使用，因此，基于该移动轨迹即可得到相应的动作对应的DS模型运动轨迹，使得可以通过上述方式，得到上述的示教运动DS模型库，且机械臂能够通过模拟人类真实的动作的方式执行任务，以便具有自主且灵活地确定运动轨迹的能力，提高了机械臂的自主性及灵活性。
可以理解的是，可以在不同场景下设置不同的示教运动DS模型库，例如，在家居场景下，可以由人类示教在家里活动的可能的动作，并基于人类示教的动作得到家居场景对应的示教运动DS模型库，在实际应用中，可以根据具体的场景及需求设置相应的示教运动DS模型库，此处不做限定。
在本发明实施例中，可以根据任务指令、环境数据及上述预设的示教运动DS模型库，确定机械臂的初始DS模型运动轨迹，且考虑到示教运动DS模型库实际上相当于一个模板库，其与机械臂真实的活动还具有差异，例如可能会存在障碍物，需要进行障碍物避障，或者可能任务的终点位置不相同等等，因此，在得到上述的初始DS模型运动轨迹之后，还将对该初始DS模型运动轨迹进行修正，得到机械臂的目标运动轨迹。
可以理解的是，在得到上述的目标运动轨迹之后，可以对机械臂进行轨迹跟踪控制，使得机械臂能够精确跟踪该目标运动轨迹，以使得机械臂能够实现上述任务指令所指示的任务，实现机械臂的类人自主运动。
在本发明实施例中，通过基于人类示教活动生成DS模型运动轨迹，并形成示教运动动态系统DS模型库的方式，使得机械臂能够利用该示教运动DS模型库确定初始DS模型运动轨迹，并进一进地对该初始DS模型运动轨迹进行修正，得到符合任务指令需求的目标运动轨迹，实现机械臂自主且灵活的运动轨迹的规划。
基于图1所示实施例，请参阅图2，为本发明实施例中机械臂的运动轨迹规划方法的另一流程示意图，包括：
步骤201、若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统 采集到的环境数据；
可以理解的是，上述步骤201与图1所示实施例中的步骤101描述的内容相似，具体可以参阅图1所示实施例中步骤101的相关内容，此处不做赘述。
步骤202、对任务指令进行解析，确定任务指令指示的目标任务名称及目标任务物体，目标任务物体是指所述机械臂执行任务指令指向的被操作物体；
步骤203、对环境数据进行物体识别，确定环境数据包含的第一物体；
步骤204、根据目标任务名称、目标任务物体，第一物体及示教运动DS模型库确定机械臂的初始DS模型运动轨迹；
在本发明实施例中，上述的示教运动DS模型库包含任务名称、任务物体及DS模型运动轨迹之间的对应关系，其中，在基于人类的示教运动得到DS模型运动轨迹时，还将设置任务物体、任务名称，例如，任务名称可以是“拿起”，任务物体可以是“水杯”、“瓶子”、“剪刀”“手机”等等，则组成的任务指令可以是：“拿起水杯”、“拿起瓶子”、“拿起剪刀”、“拿起手机”等等。可以理解的是，上述的对应关系中，任务物体可以为空，即机械臂执行的任务可以是手臂的动作，和外界的物体无关，例如，任务名称可以是“举手”、“指向前方”等等，在实际应用中，在上述对应关系中任务物体是否为空与具体的任务相关，可以根据需要进行设置，此处不做限定。
此外，上述示教运动DS模型库中的DS模型运动轨迹具体可以是：获取到人类示教活动中的针对任务A的视频数据，利用基于高斯混合模型的示教运动动态系统模型辨识方法对该视频数据进行辨识处理，得到任务A对应的DS模型运动轨迹，且可以理解的是，可以基于对任务A的分析，设置该DS模型运动轨迹的任务名称及任务物体。
在本发明实施例中，在接收到任务指令，且获取到环境数据之后，可以对该任务指令进行解析，确定该任务指令指示的目标任务名称及目标任务物体，其中，该目标任务物体是指机械臂执行任务指令指向的被操作物体，例如，若任务指令是“拿起水杯”，则目标任务名称则可以是“拿起”，目标任务物体则 是“水杯”。
具体的，若上述任务指令为非语音指令，则可以获得任务指令包含的信息，并确定该信息中的目标任务名称及目标任务物体。例如，用户可以在可与机械臂进行通信的终端的显示界面输入任务，即“打开冰箱”，并由该终端生成任务指令，且终端将该任务指令通过网络发送给机械臂，机械臂对该任务指令进行解析，则可以确定目标任务名称为“打开”，目标任务物体为“冰箱”。
若上述任务指令为语音指令，则需要先对该语音指令进行语音识别，确定该语音指令包含的文字内容，并进一步地对文字内容进行关键字划分，确定关键字，然后基于关键字确定上述的目标任务名称及目标任务物体。
进一步地，还将对环境数据进行物体识别，确定该环境数据包含的第一物体，其中，上述的第一物体是指环境数据包含的图像中存在的所有物体。
且在一种可行的实现方式中，将根据上述的目标任务名称、目标任务物体，第一物体及示教运动DS模型库确定机械臂的初始DS模型运动轨迹，其中，由于示教运动DS模型库包含任务名称、任务物体及DS模型运动轨迹，因此，上述初始DS模型运动轨迹具体可以按照如下方式得到：
步骤a、判断第一物体是否包含目标任务物体；
步骤b、当第一物体包含目标任务物体时，利用目标任务名称、目标任务物体查找示教运动DS模型库，确定与目标任务名称及目标任务物体具有对应关系的初始DS模型运动轨迹。
其中，将判断第一物体是否包含目标任务物体，具体的，若目标任务物体不为空，则遍历第一物体，且遍历结束后，确定遍历到的物体中与目标任务物体相同的第二物体，则确定该第二物体即为目标任务物体，例如，若第一物体包括：水杯、剪刀、笔、本子等，目标任务物体为“水杯”，则遍历第一物体确定包含目标任务物体“水杯”。可以理解的是，在目标任务物体为空时，则也可以确定第一物体包含目标任务物体。
进一步地，将利用目标任务名称、目标任务物体查找示教运动DS模型库， 确定与目标任务名称及目标任务物体具有对应关系的初始DS模型运动轨迹。
可以理解的是，任务指令指示的任务可以是由多个小的任务构成的，例如，任务是“打开冰箱”，则可以将该任务分解成两个步骤执行，一个为移动到冰箱门处，其对应的目标任务名称是：移动位置，目标任务物体为：冰箱，另一个则是执行打开的动作，其对应的目标任务名称为：打开，目标任务物体为冰箱，在该种情况下，则需要分别得到该两个步骤对应的初始DS模型运动轨迹，将该两个步骤的初始DS模型运动轨迹作为任务指令指示的任务的运动轨迹，可以理解的是，在实际应用中可以根据需要设置对任务的划分方式，及示教运动DS模型库中各个DS模型运动轨迹对应的任务大小，此处不做限定。
步骤205、对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
在本发明实施例中，通过使用目标任务名称及目标任务物体查找示教运动DS模型库的方式，使得能够有效的确定与目标任务名称及目标任务物体具有对应关系的初始DS模型运动轨迹。
进一步地，请参阅图3，为本发明实施例中图1所示实施例中步骤103的细化步骤的流程示意图，该细化步骤包括：
步骤301、判断环境数据包含的第一物体中是否存在位姿位于初始DS模型运动轨迹上的障碍物，得到第一判断结果；及判断任务指令中的目标任务物体的第一位姿是否与初始DS模型运动轨迹中的目标物体的第二位姿相同，得到第二判断结果；
步骤302、根据第一判断结果及第二判断结果对初始DS模型运动轨迹进行修正，得到目标运动轨迹。
在本发明实施例中，在得到初始DS模型运动轨迹之后，将对其进行修正，使得修正后得到的目标运动轨迹能够更符合实际场景下机械臂执行任务指令所需要的轨迹。
具体的，可以先确定环境数据包含的第一物体中是否存在位姿位于初始 DS模型运动轨迹上的障碍物，使得能够进行障碍物的判断，以便确定是否需要进行避障处理。
其中，上述的环境数据包含的第一物体，是通过对环境数据进行物体识别确定的，除了用于确定是否存在目标任务物体之外，还用于确定是否存在障碍物，可以理解的是，机械臂可以确定环境数据中包含的第一物体的位姿，该位姿及DS模型运动轨迹都是基于同一个笛卡尔坐标系确定的。
其中，确定障碍物的方式可以是判断第一物体中是否存在位姿位于初始DS模型运动轨迹上的障碍物，得到第一判断结果，该第一判断结果可以是不存在，或者，可以是存在，且为物体A，通过该种方式，可以确定是否存在障碍物，可以理解的是，确定障碍物可以是实时的，即实时确定在机械臂的末端的移动轨迹上，该末端即将达到的位置是否存在障碍物。
此外，还将确定目标任务物体的第一位姿是否与初始DS模型运动轨迹中的目标物体的第二位姿相同，得到第二判断结果，该第二判断结果可以是相同也可以不同。其中，初始DS模型运动轨迹中的目标物体是指人类示教活动所使用到的物体，例如，人类示教活动是“拿起水杯”，则目标物体则为水杯。
具体的，若基于第一判断结果确定存在障碍物，且基于第二判断结果确定第一位姿与第二位姿相同，则对初始DS模型运动轨迹进行障碍物避障修正，得到目标运动轨迹。
若基于第一判断结果确定存在障碍物，且基于第二判断结果确定第一位姿与第二位姿不同，则对初始DS模型运动轨迹进行障碍物避障修正，并基于第一位姿对障碍物避障修正后的初始DS模型运动轨迹进行目标物体的位姿的修正，得到目标运动轨迹；
若基于第一判断结果确定不存在障碍物，且基于第二判断结果确定第一位姿与第二位姿相同，则确定初始DS模型运动轨迹为目标运动轨迹；
若基于第一判断结果确定不存在障碍物，且基于第二判断结果确定第一位姿与第二位姿不同，则基于第一位姿对初始DS模型运动轨迹进行目标物体的 位姿的修正，得到目标运动轨迹。
在本发明实施例中，上述的对初始DS模型运动轨迹进行障碍物避障修正，包括：先基于障碍物的形状，确定障碍物为凹形障碍物还是凸形障碍物，其中，凹形障碍物是指物体的形状的外边缘的切线可将其形状划分成至少两个部分的物体，除了凹形障碍物以外的其他障碍物都可以称为凸形障碍物。
其中，当障碍物为凹形障碍物时，可以将凹形障碍物划分成至少两个两两相交的凸形障碍物，其中，两两相交的凸形障碍物相交的位置为相交线。
当初始DS模型运动轨迹到达凹形障碍物的交点未位于相交线上时，计算凸形障碍物组合形成的组合模态矩阵，利用组合模态矩阵对初始DS模型运动轨迹进行修正。
当初始DS模型运动轨迹到达凹形障碍物的交点位于相交线上时，确定形成相交线的第一凸形障碍物的第一法向量及第二凸形障碍物的第二法向量，利用第一法向量及第二法向量构建交点对应的模态矩阵，利用交点对应的模态矩阵对初始DS模型运动轨迹进行修正。
为了更好的理解本发明实施例中的技术方案，请参阅图4，为本发明实施例中凹形障碍物的一个示意图，该凹形障碍物是以电钻为例，其由三个凸形障碍物构成，该三个凸形障碍物分别为凸形障碍物1、凸形障碍物2及凸形障碍物3，其中，凸形障碍物1和凸形障碍物2之间两两相交，凸形障碍物2和凸形障碍物3之间两两相交，且两两相交的凸形障碍物的相交的位置为相交线，在图4中，相交线为A和B，其中，相交线A的范围为点c1至点c2之间，相交线B的范围为点c3至点c4之间。可以理解的是，对于任何一个凹形障碍物，都可以划分成多个凸形障碍物。
可以按照如下公式计算上述电钻划分后的三个凸形障碍物的组合模态矩阵，如下：
Figure PCTCN2020135084-appb-000001
其中，
Figure PCTCN2020135084-appb-000002
Figure PCTCN2020135084-appb-000003
Figure PCTCN2020135084-appb-000004
其中，
Figure PCTCN2020135084-appb-000005
表示N个凸形障碍物的组合模态矩阵，
Figure PCTCN2020135084-appb-000006
表示第i个凸形障碍物的模态矩阵，
Figure PCTCN2020135084-appb-000007
表示第i个凸形障碍物的法向量，
Figure PCTCN2020135084-appb-000008
和
Figure PCTCN2020135084-appb-000009
为第i个凸形障碍物的法向量对应的超平面的基向量；
其中，
Figure PCTCN2020135084-appb-000010
其中，
Figure PCTCN2020135084-appb-000011
表示第i个凸形障碍物的表面函数，(ξ) 1、(ξ) 2、(ξ) 3分别表示笛卡尔坐标系中的X轴、Y轴及Z轴。
在得到上述凹形障碍物拆分后的多个凸形障碍物的组合模态之后，将进一步地确定初始DS模型运动轨迹到达该凹形障碍物的交点是否位于相交线上，即如图4所示，确定初始DS模型运动轨迹是否到达了相交线A或者相交线B上。
当初始模型运动轨迹到达凹形障碍物的交点未位于相交线上时，此时，可利用上述的组合模态矩阵对初始DS模型运动轨迹进行修正。
具体的，可以使用如下公式：
Figure PCTCN2020135084-appb-000012
其中，
Figure PCTCN2020135084-appb-000013
表示初始DS模型运动轨迹进行障碍物修正之后的运动轨迹，
Figure PCTCN2020135084-appb-000014
表示N个凸形障碍物的组合模态矩阵，f(ξ)表示初始DS模型运动轨迹。
具体的，以图4所涉及的内容为基础，请参阅图5，为图4所示凹形障碍物的另一示意图，其中，
Figure PCTCN2020135084-appb-000015
表示凸形障碍物1的中心点，
Figure PCTCN2020135084-appb-000016
表示凸形障碍物2的中心点，
Figure PCTCN2020135084-appb-000017
表示凸形障碍物3的中心点，
Figure PCTCN2020135084-appb-000018
表示交点，且该交点位于第三凸形障碍物的表面，
Figure PCTCN2020135084-appb-000019
表示第三个凸形障碍物的法向量，
Figure PCTCN2020135084-appb-000020
和
Figure PCTCN2020135084-appb-000021
表 示第三个凸形障碍物的法向量对应的超平面的基向量。
当初始DS模型运动轨迹到达凹形障碍物的交点位于相交线上时，则确定形成相交线的第一凸形障碍物的第一法向量及第二凸形障碍物的第二法向量，利用第一法向量及第二法向量构建交点对应的模态矩阵，利用该交点对应的模态矩阵对初始DS模型运动轨迹进行修正，使得能够避开凹形障碍物。
其中，以交点为相交线上的点
Figure PCTCN2020135084-appb-000022
为例，且点
Figure PCTCN2020135084-appb-000023
在第m个凸形障碍物和第n个凸形障碍物的相交线上，则点
Figure PCTCN2020135084-appb-000024
在第m个凸形障碍物的第一法向量(在第一凸形障碍物的第一法向量)如下：
Figure PCTCN2020135084-appb-000025
Figure PCTCN2020135084-appb-000026
其中，
Figure PCTCN2020135084-appb-000027
表示第一法向量，
Figure PCTCN2020135084-appb-000028
表示交点的坐标，
Figure PCTCN2020135084-appb-000029
表示第m个凸形障碍物的中心点的坐标，
Figure PCTCN2020135084-appb-000030
表示第m个凸形障碍物的表面函数，b代表第m个凸形障碍物和第n个凸形障碍物交线上一点的标记。
其中，点
Figure PCTCN2020135084-appb-000031
在第n个凸形障碍物的第二法向量(在第二凸形障碍物的第二法向量)如下：
Figure PCTCN2020135084-appb-000032
Figure PCTCN2020135084-appb-000033
其中，
Figure PCTCN2020135084-appb-000034
表示第二法向量，
Figure PCTCN2020135084-appb-000035
表示交点的坐标，
Figure PCTCN2020135084-appb-000036
表示第m个凸形障碍物的中心点的坐标，
Figure PCTCN2020135084-appb-000037
表示第n个凸形障碍物的表面函数，b代表第m个凸形障碍物和第n个凸形障碍物交线上一点的标记。
在得到上述的第一法向量和第二法向量之后，利用第一法向量和第二法向量构建交点
Figure PCTCN2020135084-appb-000038
对应的
Figure PCTCN2020135084-appb-000039
模态矩阵，如下：
Figure PCTCN2020135084-appb-000040
其中，
Figure PCTCN2020135084-appb-000041
Figure PCTCN2020135084-appb-000042
Figure PCTCN2020135084-appb-000043
Figure PCTCN2020135084-appb-000044
其中，pin表示伪逆。
其中，
Figure PCTCN2020135084-appb-000045
表示与第一法向量和第二法向量均垂直的向量。通过构造该向量的方式，使得在对初始移动轨迹进行避障修正时，能够使得修正后的目标移动轨迹能够沿着第一障碍物和第二障碍物的表面的法向量的切线向外移出，以实现避障。
且在得到交点
Figure PCTCN2020135084-appb-000046
对应的
Figure PCTCN2020135084-appb-000047
模态矩阵，将利用该模态矩阵对初始DS模型运动轨迹进行修正，得到进行障碍物修正后的运动轨迹。公式如下：
Figure PCTCN2020135084-appb-000048
其中，f(ξ)表示初始移动轨迹，
Figure PCTCN2020135084-appb-000049
表示基于相交点
Figure PCTCN2020135084-appb-000050
确定的目标模态矩阵，
Figure PCTCN2020135084-appb-000051
表示避障修正之后的目标移动轨迹。
具体的，以图4所涉及的内容为基础，请参阅图5，为图4所示凹形障碍物的另一示意图，其中，
Figure PCTCN2020135084-appb-000052
表示凸形障碍物1的中心点，
Figure PCTCN2020135084-appb-000053
表示凸形障碍物2的中心点，
Figure PCTCN2020135084-appb-000054
表示凸形障碍物3的中心点，
Figure PCTCN2020135084-appb-000055
表示第一凸形障碍物1和第二凸形障碍物2的相交线(交线)上的点，且该点为上述的交点。
Figure PCTCN2020135084-appb-000056
表示点
Figure PCTCN2020135084-appb-000057
第一个凸形障碍物上的法向量，
Figure PCTCN2020135084-appb-000058
表示点
Figure PCTCN2020135084-appb-000059
第二个凸形障碍物上的法向量，e 12表示同时垂直于
Figure PCTCN2020135084-appb-000060
和
Figure PCTCN2020135084-appb-000061
的向量，即上述的
Figure PCTCN2020135084-appb-000062
的值。
可以理解的是，若障碍物修正之后不需要再进行目标物体的位姿的修正，则障碍物修正后的运动轨迹即为目标运动轨迹，若障碍物修正之后还需要对目 标物体的位姿进行修正，则进一步的进行目标物体的位姿的，且将修正后的运动轨迹作为目标运动轨迹。
其中，目标物体的位姿的修正方式为利用所述目标任务物体的第一位姿替换所述目标物体的位姿。
在本发明实施例中，通过利用环境数据及任务指令，可以有效确定是否需要进行障碍物避障修正及目标任务物体的位姿的修正，且在需要进行障碍物避障修正时，若障碍物为凹形障碍物，则通过将凹形障碍物划分成多个凸形障碍物，并基于初始DS模型运动轨迹到达凹形障碍物的交点是否位于相交线上，而构造不同的模态矩阵，以对初始DS模型运动轨迹进行修正，使得能够有效实现避障，且在需要对目标物体的位姿进行修正时，也将进行目标物体的位姿的修正，通过上述方式，使得修正后的目标运动轨迹更加符合实际场景下的任务需求，使得能够有效且精确的实现任务。
且进一步地，本发明实施例中的技术方案适用于诸如养老、助残、家庭服务、人机协同等等场景，能够使得机械臂自主且灵活的实现运动轨迹的规划，且通过对确定的目标运动轨迹的跟踪控制，使得机械臂能够实现精确控制，有效且准确地完成期望的任务，实现机械臂的类人自主运动。且可以理解的是，上述的运动轨迹的规划过程是实时的过程，也可以实现运动轨迹的实时规划。
图6示出了一个实施例中机械臂的系统方框图的示意图。如图6所示，该机械臂包括通过系统总线连接的处理器、存储器和网络接口。其中，存储器包括非易失性存储介质和内存储器。该机械臂的非易失性存储介质存储有操作系统，还可存储有计算机程序，该计算机程序被处理器执行时，可使得处理器实现机械臂的运动轨迹规划方法。该内存储器中也可储存有计算机程序，该计算机程序被处理器执行时，可使得处理器执行机械臂的运动轨迹规划方法。本领域技术人员可以理解，图6中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的机械臂的限定，具体的机械臂可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同 的部件布置。
在一个实施例中，提出一种机械臂的运动轨迹规划装置，装置包括：处理器及存储器，存储器中存储有计算机程序，处理器执行计算机程序时，使得处理器执行以下步骤：
若机械臂接收到任务指令，则获取机械臂的视觉检测系统采集到的环境数据；
根据任务指令、环境数据及预设的示教运动动态系统DS模型库，确定机械臂的初始DS模型运动轨迹，示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
对初始DS模型运动轨迹进行修正，得到机械臂的目标运动轨迹。
在一个实施例中，提出了一种机械臂，包括存储器和处理器，所述存储器存储有计算机程序，所述计算机程序被所述处理器执行时，使得所述处理器执行以下步骤：
若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
在一个实施例中，提出了一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时，使得所述处理器执行以下步骤：
若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
在本发明实施例中，通过基于人类示教活动生成DS模型运动轨迹，并形成示教运动动态系统DS模型库的方式，使得机械臂能够利用该示教运动DS模型库确定初始DS模型运动轨迹，并进一进地对该初始DS模型运动轨迹进行修正，得到符合任务指令需求的目标运动轨迹，实现机械臂自主且灵活的运动轨迹的规划。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一非易失性计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对本申请专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (10)

  1. 一种机械臂的运动轨迹规划方法，其特征在于，所述方法包括：
    若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
    根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
    对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，包括：
    对所述任务指令进行解析，确定所述任务指令指示的目标任务名称及目标任务物体，所述目标任务物体是指所述机械臂执行所述任务指令指向的被操作物体；
    对所述环境数据进行物体识别，确定所述环境数据包含的第一物体；
    根据所述目标任务名称、所述目标任务物体，所述第一物体及所述示教运动DS模型库确定所述机械臂的初始DS模型运动轨迹。
  3. 根据权利要求2所述的方法，其特征在于，所述示教运动DS模型库包含任务名称、任务物体及DS模型运动轨迹之间的对应关系；
    则所述根据所述目标任务名称、所述目标任务物体，所述第一物体及所述示教运动DS模型库确定所述机械臂的初始DS模型运动轨迹，包括：
    判断所述第一物体是否包含所述目标任务物体；
    当所述第一物体包含所述目标任务物体时，利用所述目标任务名称、所述目标任务物体查找所述示教运动DS模型库，确定与所述目标任务名称及所述目标任务物体具有对应关系的所述初始DS模型运动轨迹。
  4. 根据权利要求1所述的方法，其特征在于，所述对所述DS模型运动 轨迹进行修正，得到所述机械臂的目标运动轨迹，包括：
    判断所述环境数据包含的第一物体中是否存在位姿位于所述初始DS模型运动轨迹上的障碍物，得到第一判断结果；及判断所述任务指令中的目标任务物体的第一位姿是否与所述初始DS模型运动轨迹中的目标物体的第二位姿相同，得到第二判断结果；
    根据所述第一判断结果及所述第二判断结果对所述初始DS模型运动轨迹进行修正，得到所述目标运动轨迹。
  5. 根据权利要求4所述的方法，其特征在于，所述根据所述第一判断结果及所述第二判断结果对所述初始DS模型运动轨迹进行修正，得到所述目标运动轨迹，包括：
    若存在所述障碍物，且所述第一位姿与所述第二位姿相同，则对所述初始DS模型运动轨迹进行障碍物避障修正，得到所述目标运动轨迹；
    若存在所述障碍物，且所述第一位姿与所述第二位姿不同，则对所述初始DS模型运动轨迹进行障碍物避障修正，并基于所述第一位姿对障碍物避障修正后的初始DS模型运动轨迹进行所述目标物体的位姿的修正，得到所述目标运动轨迹；
    若不存在所述障碍物，且所述第一位姿与所述第二位姿相同，则确定所述初始DS模型运动轨迹为所述目标运动轨迹；
    若不存在所述障碍物，且所述第一位姿与所述第二位姿不同，则基于所述第一位姿对所述初始DS模型运动轨迹进行所述目标点的位姿的修正，得到所述目标运动轨迹。
  6. 根据权利要求5所述的方法，其特征在于，所述对所述初始DS模型运动轨迹进行障碍物避障修正，包括：
    当所述障碍物为凹形障碍物时，将所述凹形障碍物划分成至少两个两两相交的凸形障碍物，其中，两两相交的凸形障碍物相交的位置为相交线；
    当所述初始DS模型运动轨迹到达所述凹形障碍物的交点未位于所述相交 线上时，计算所述凸形障碍物组合形成的组合模态矩阵，利用所述组合模态矩阵对所述初始DS模型运动轨迹进行修正。
  7. 根据权利要求6所述的方法，其特征在于，所述方法还包括：
    当所述初始DS模型运动轨迹到达所述凹形障碍物的交点位于所述相交线上时，确定形成所述相交线的第一凸形障碍物的第一法向量及第二凸形障碍物的第二法向量；
    利用所述第一法向量及所述第二法向量构建所述交点对应的模态矩阵；
    利用所述交点对应的模态矩阵对所述初始DS模型运动轨迹进行修正。
  8. 一种机械臂的运动轨迹规划装置，其特征在于，所述装置包括：处理器及存储器，所述存储器中存储有计算机程序，所述处理器执行所述计算机程序时，使得所述处理器执行以下步骤：
    若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
    根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
    对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
  9. 一种计算机可读存储介质，存储有计算机程序，其特征在于，所述计算机程序被处理器执行时，使得所述处理器执行以下步骤：
    若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
    根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
    对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
  10. 一种机械臂，包括存储器和处理器，其特征在于，所述存储器存储有 计算机程序，所述计算机程序被所述处理器执行时，使得所述处理器执行以下步骤：
    若机械臂接收到任务指令，则获取所述机械臂的视觉检测系统采集到的环境数据；
    根据所述任务指令、所述环境数据及预设的示教运动动态系统DS模型库，确定所述机械臂的初始DS模型运动轨迹，所述示教运动DS模型库至少包含基于人类示教活动生成的DS模型运动轨迹；
    对所述初始DS模型运动轨迹进行修正，得到所述机械臂的目标运动轨迹。
PCT/CN2020/135084 2020-12-10 2020-12-10 机械臂的运动轨迹规划方法及装置、机械臂及存储介质 WO2022120670A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2020/135084 WO2022120670A1 (zh) 2020-12-10 2020-12-10 机械臂的运动轨迹规划方法及装置、机械臂及存储介质
US17/566,726 US20220184808A1 (en) 2020-12-10 2021-12-31 Motion trajectory planning method for robotic manipulator, robotic manipulator and computer-readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/135084 WO2022120670A1 (zh) 2020-12-10 2020-12-10 机械臂的运动轨迹规划方法及装置、机械臂及存储介质

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/566,726 Continuation US20220184808A1 (en) 2020-12-10 2021-12-31 Motion trajectory planning method for robotic manipulator, robotic manipulator and computer-readable storage medium

Publications (1)

Publication Number Publication Date
WO2022120670A1 true WO2022120670A1 (zh) 2022-06-16

Family

ID=81943024

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/135084 WO2022120670A1 (zh) 2020-12-10 2020-12-10 机械臂的运动轨迹规划方法及装置、机械臂及存储介质

Country Status (2)

Country Link
US (1) US20220184808A1 (zh)
WO (1) WO2022120670A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116985123A (zh) * 2023-07-17 2023-11-03 华东交通大学 一种降低长柔性液压机械臂振动的轨迹规划方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112578795A (zh) * 2020-12-15 2021-03-30 深圳市优必选科技股份有限公司 机器人避障方法及装置、机器人及存储介质
US20220402140A1 (en) * 2021-06-18 2022-12-22 Intrinsic Innovation Llc Learning to acquire and adapt contact-rich manipulation skills with motion primitives

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120165982A1 (en) * 2010-12-27 2012-06-28 Samsung Electronics Co., Ltd. Apparatus for planning path of robot and method thereof
CN106909216A (zh) * 2017-01-05 2017-06-30 华南理工大学 一种基于Kinect传感器的仿人机械手控制方法
CN107214701A (zh) * 2017-06-12 2017-09-29 南京理工大学 一种基于运动基元库的带电作业机械臂自主避障路径规划方法
CN108237534A (zh) * 2018-01-04 2018-07-03 清华大学深圳研究生院 一种连续型机械臂的空间避障轨迹规划方法

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9486918B1 (en) * 2013-03-13 2016-11-08 Hrl Laboratories, Llc System and method for quick scripting of tasks for autonomous robotic manipulation
US10788836B2 (en) * 2016-02-29 2020-09-29 AI Incorporated Obstacle recognition method for autonomous robots
US11016491B1 (en) * 2018-01-26 2021-05-25 X Development Llc Trajectory planning for mobile robots
US11458626B2 (en) * 2018-02-05 2022-10-04 Canon Kabushiki Kaisha Trajectory generating method, and trajectory generating apparatus
US11597084B2 (en) * 2018-09-13 2023-03-07 The Charles Stark Draper Laboratory, Inc. Controlling robot torque and velocity based on context
WO2020106706A1 (en) * 2018-11-19 2020-05-28 Siemens Aktiengesellschaft Object marking to support tasks by autonomous machines
US11731271B2 (en) * 2020-06-30 2023-08-22 Microsoft Technology Licensing, Llc Verbal-based focus-of-attention task model encoder

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120165982A1 (en) * 2010-12-27 2012-06-28 Samsung Electronics Co., Ltd. Apparatus for planning path of robot and method thereof
CN106909216A (zh) * 2017-01-05 2017-06-30 华南理工大学 一种基于Kinect传感器的仿人机械手控制方法
CN107214701A (zh) * 2017-06-12 2017-09-29 南京理工大学 一种基于运动基元库的带电作业机械臂自主避障路径规划方法
CN108237534A (zh) * 2018-01-04 2018-07-03 清华大学深圳研究生院 一种连续型机械臂的空间避障轨迹规划方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YU JIANJUN, ET AL.: "Research on Robot Imitation Learning Method based on Dynamical System", CAAI TRANSACTIONS ON INTELLIGENT SYSTEMS, vol. 14, no. 5, 30 September 2019 (2019-09-30), pages 1026 - 1034, XP055941741, ISSN: 1673-4785, DOI: 10.11992/tis.201807018 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116985123A (zh) * 2023-07-17 2023-11-03 华东交通大学 一种降低长柔性液压机械臂振动的轨迹规划方法

Also Published As

Publication number Publication date
US20220184808A1 (en) 2022-06-16

Similar Documents

Publication Publication Date Title
WO2022120670A1 (zh) 机械臂的运动轨迹规划方法及装置、机械臂及存储介质
Pérez-D'Arpino et al. C-learn: Learning geometric constraints from demonstrations for multi-step manipulation in shared autonomy
CN112518756B (zh) 机械臂的运动轨迹规划方法及装置、机械臂及存储介质
WO2020168770A1 (zh) 物体位姿估计方法及装置
Kaltsoukalas et al. On generating the motion of industrial robot manipulators
Sundaralingam et al. Geometric in-hand regrasp planning: Alternating optimization of finger gaits and in-grasp manipulation
Ye et al. Collision-free motion planning for the litchi-picking robot
Wang et al. Collision-free trajectory planning in human-robot interaction through hand movement prediction from vision
Dang et al. Semantic grasping: planning task-specific stable robotic grasps
WO2020190166A1 (ru) Способ и система захвата объекта с помощью роботизированного устройства
Klingensmith et al. Closed-loop servoing using real-time markerless arm tracking
Nogueira et al. Self-adaptive cobots in cyber-physical production systems
Jha et al. Robot programming by demonstration using teleoperation through imitation
Tarbouriech et al. Bi-objective motion planning approach for safe motions: Application to a collaborative robot
CN112917470A (zh) 一种机械手的示教方法、装置、系统、存储介质及设备
Zhang et al. Subtask-learning based for robot self-assembly in flexible collaborative assembly in manufacturing
Palm et al. Recognition of human grasps by time-clustering and fuzzy modeling
WO2022126352A1 (zh) 机器人避障方法及装置、机器人及存储介质
Ku et al. Modeling objects as aspect transition graphs to support manipulation
JP7531733B2 (ja) 異なるドメイン内のタスク間の転送
CN116476063A (zh) 一种机器人操作任务实例化方法及相关装置
Das et al. Stochastic modeling of distance to collision for robot manipulators
Pedrosa et al. A skill-based architecture for pick and place manipulation tasks
Fu et al. Robotic arm intelligent grasping system for garbage recycling
Becerra et al. Smart manipulation approach for assistant robot

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: 20964609

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: 20964609

Country of ref document: EP

Kind code of ref document: A1