Robot motion brain design method and device, electronic equipment and robot

Abstract

The invention discloses a robot motion brain design method and device, electronic equipment and a robot. The method comprises the following steps: acquiring internal state information of the robot based on a first sensor system on the robot, and acquiring environment information of a target environment where the robot is located based on a second sensor system on the robot; performing three-dimensional modeling on the robot and the target environment based on the internal state information and the environment information to obtain a three-dimensional attitude model corresponding to the robot and a three-dimensional environment model corresponding to the target environment; determining a contact point between the robot and the target environment based on the three-dimensional attitude model and the three-dimensional environment model; performing motion planning on the robot based on the contact point to obtain a motion planning result; and controlling the robot to move based on the motion planning result. According to the invention, the technical problems of inaccurate path planning and flexible operation and poor balance of the robot in the motion process caused by incomplete consideration of factors during robot motion control in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of robots, and in particular to a method and device for designing a robot motor brain, an electronic device and a robot. Background Art

[0002] With the rapid development of robotics, especially the breakthroughs in brain-like intelligence and intelligent perception, control, and decision-making of robots, people's expectations for robots have expanded from simple pre-programmed execution to the ability to autonomously complete tasks in complex and ever-changing environments. In this context, the design and control of multi-joint mobile robots face unprecedented challenges, especially humanoid robots and composite robots with multiple legs and arms.

[0003] In the related art, the control of multi-joint mobile robots usually relies on a wheeled chassis or a tracked mobile platform, combined with a robotic arm to perform operational tasks. Although this composite robot solution has achieved the ability of the robot to move in the environment and complete tasks to a certain extent, it has obvious limitations. First, the external perception and control of the robotic arm in the composite robot are independent of each other, that is, the functions are jointly realized through the combination of two controllers, lacking an overall, highly integrated control architecture. Secondly, during the movement of the composite robot, the robotic arm often maintains a fixed posture, which means that the operational flexibility and adaptability in a dynamic environment are limited. In addition, multi-joint robots often need to physically interact with the environment when performing tasks, such as grasping objects, walking or manipulating tools, which places strict requirements on the robot's balance control and posture adjustment. Especially on uneven terrain, or in situations where it interacts with multiple objects, the robot must be able to accurately predict and adjust its center of mass position to maintain dynamic balance and avoid falling or collisions.

[0004] Although the design methods of multi-joint mobile robot motion brains in related technologies perform well in certain specific scenarios, they are insufficient in terms of dynamic environment perception, real-time posture adjustment, high-precision balance maintenance, and efficient task completion. For example, they may not be able to respond quickly to environmental changes, such as sudden obstacles or changes in the position of target objects, or they may consume too much energy when performing tasks, or be inefficient due to insufficient operation strategies.

[0005] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0006] The embodiments of the present invention provide a robot motion brain design method, device, electronic device and robot, so as to at least solve the technical problems in the related art that factors are not fully considered when performing robot motion control, resulting in inaccurate path planning and dexterous operation, and poor balance of the robot during movement.

[0007] According to one aspect of an embodiment of the present invention, a method for designing a robot motion brain is provided, comprising: collecting internal state information of the robot based on a first sensor system on the robot, and collecting environmental information of a target environment in which the robot is located based on a second sensor system on the robot; performing three-dimensional modeling on the robot and the target environment based on the internal state information and the environmental information, respectively, to obtain a three-dimensional posture model corresponding to the robot, and a three-dimensional environment model corresponding to the target environment; determining contact points between the robot and the target environment based on the three-dimensional posture model and the three-dimensional environment model; performing motion planning on the robot based on the contact points to obtain motion planning results; and controlling the movement of the robot based on the motion planning results.

[0008] According to another aspect of an embodiment of the present invention, a robot motion brain design device is also provided, including: an information acquisition module, used to acquire internal state information of the robot based on a first sensor system on the robot, and to acquire environmental information of the target environment in which the robot is located based on a second sensor system on the robot; a three-dimensional modeling module, used to perform three-dimensional modeling of the robot and the target environment respectively based on the internal state information and the environmental information, and obtain a three-dimensional posture model corresponding to the robot, and a three-dimensional environment model corresponding to the target environment; a contact point determination module, used to determine the contact points between the robot and the target environment based on the three-dimensional posture model and the three-dimensional environment model; a motion planning module, used to perform motion planning for the robot based on the contact points, and obtain motion planning results; and a motion control module, used to control the movement of the robot based on the motion planning results.

[0009] According to another aspect of an embodiment of the present invention, a robot is provided, comprising: a robot body, on which a first sensor system and a second sensor system are arranged; and a controller connected to the robot body, for executing any one of the robot motor brain design methods described.

[0010] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any one of the above-mentioned robot motion brain design methods.

[0011] In an embodiment of the present invention, the internal state information of the robot is collected based on a first sensor system on the robot, and the environmental information of the target environment of the robot is collected based on a second sensor system on the robot; based on the internal state information and the environmental information, three-dimensional modeling is performed on the robot and the target environment respectively, to obtain a three-dimensional posture model corresponding to the robot, and a three-dimensional environment model corresponding to the target environment; based on the three-dimensional posture model and the three-dimensional environment model, the contact points between the robot and the target environment are determined; based on the contact points, motion planning is performed on the robot to obtain motion planning results; based on the motion planning results, the robot motion is controlled, thereby achieving the purpose of comprehensive three-dimensional modeling and robot motion planning based on internal state information and environmental information, thereby realizing the technical effect of improving the accuracy of robot path planning and operation balance, and further solving the technical problem in the related technology that factors are not fully considered when performing robot motion control, resulting in inaccurate path planning and dexterous operation, and poor balance of the robot during movement. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0013] Figure 1 is a schematic diagram of a method for designing a robot motion brain according to an embodiment of the present invention;

[0014] Figure 2 is a schematic diagram of an optional robot motion brain design method according to an embodiment of the present invention;

[0015] Figure 3 Schematic diagram of a robot motion brain design device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0017] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0018] First, to facilitate understanding of the embodiments of the present invention, some terms or nouns involved in the present invention are explained below:

[0019] The robot motion brain refers to a decision-making system that enables the robot to have autonomous movement capabilities based on the robot's morphology and combines forward and inverse kinematics, dynamics and artificial intelligence technologies.

[0020] According to an embodiment of the present invention, a method embodiment of a robot motor brain design is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0021] Figure 1 is a flow chart of a method for designing a robot motion brain according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0022] Step S102, collecting internal state information of the robot based on a first sensor system on the robot, and collecting environmental information of a target environment where the robot is located based on a second sensor system on the robot;

[0023] In the above step S102, the first sensor system and the second sensor system respectively collect the internal state information of the robot and the environmental information of the target environment, so that the robot can perceive the changes of its own state and the surrounding environment in real time, thereby improving its adaptability and operation accuracy in complex environments.

[0024] Optionally, the robot can be but is not limited to a multi-joint mobile robot. The first sensor system can be understood as the robot's internal perception system, through which the speed, position and force information of all joints of the robot can be perceived, the center of mass position of the robot can be evaluated, the real-time posture of the robot can be preliminarily identified, and the internal state information of the robot can be obtained. The internal perception system can include but is not limited to encoders, force / torque sensors, current sensors, voltage / remaining power detection sensors, temperature sensors and other sensors, wherein the encoder is used to determine the rotation speed of each degree of freedom of the robot, the force / torque sensor is used to detect the force or torque output by the end of the robot's limbs, and the output torque of the corresponding joint or degree of freedom of the robot can be calculated by the joint current value, and the voltage / remaining power detection sensor is used to calculate the robot's endurance time; the temperature is used to determine the working state of the robot's internal controller, battery, etc.

[0025] Optionally, the second sensor system can be understood as an external perception system, which is used to perceive the robot's surrounding environment information, measure the distance from all objects around the robot to the robot, identify and locate the target object, and obtain the environmental information of the robot's target environment. The external perception system may include, but is not limited to, a lidar sensor, a camera, a voice dialogue device, an infrared spectrum sensor, an ultrasonic sensor, and other sensors, wherein the relative coordinates between each sensor are established based on the relative position relationship (distance, angle) of each sensor distributed on the robot; the features of the same external object are used as a reference; and the three-dimensional point cloud models of each sensor are fused. Voice dialogue belongs to the external sensing system, and its main function is to interact with people and understand the intention of human voice.

[0026] Step S104, based on the internal state information and the environmental information, three-dimensional modeling is performed on the robot and the target environment respectively to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment;

[0027] In the above step S104, based on the collected internal state information and environmental information, three-dimensional modeling of the robot and the target environment is performed, which helps the robot to more accurately understand and analyze its position, posture and relative relationship with the target object in the environment, so as to make more reasonable decisions.

[0028] In an optional embodiment, based on internal state information and environmental information, three-dimensional models are performed on the robot and the target environment respectively to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment, including: during the movement of the robot, based on internal state information and environmental information, three-dimensional models are performed on the robot and the target environment respectively to obtain a three-dimensional posture model and a three-dimensional environment model.

[0029] Optionally, continuous 3D modeling during robot motion means that the robot can perceive and understand changes in its surrounding environment and its own posture in real time. This dynamic perception is essential for maintaining stable and efficient task execution in a changing environment. Through real-time modeling, the robot can quickly update its 3D posture model in the world coordinate system and the 3D environmental model of the target environment, ensuring the robot's real-time cognition and control of itself and the environment, improving its reaction speed and control accuracy. Continuous 3D modeling allows the robot to dynamically adjust its task planning during motion, such as path selection, obstacle avoidance strategy or operation strategy, ensuring that its actions are not only efficient, but also able to adapt to environmental changes and improve the success rate of task completion. Real-time updating of posture and environmental models during movement helps the robot dynamically maintain balance. For multi-joint robots, especially humanoid robots, being able to stably walk or operate on complex terrain or unstable surfaces can significantly improve the safety of robot operation. Dynamic modeling can reduce the risk of misoperation caused by environmental changes or sensor data delays. The robot can adjust its actions in time to avoid unnecessary collisions or interference with objects in the environment.

[0030] Step S106, determining the contact point between the robot and the target environment based on the three-dimensional posture model and the three-dimensional environment model;

[0031] In the above step S106, the contact points between the robot and the target environment are determined by analyzing the three-dimensional posture model and the three-dimensional environment model. This step is crucial for the robot to be able to stably interact with objects when performing tasks, especially when it needs to maintain balance or operate on an unstable surface.

[0032] In an optional embodiment, the contact point between the robot and the target environment is determined based on the three-dimensional posture model and the three-dimensional environment model, including: unifying the three-dimensional posture model and the three-dimensional environment model into the same coordinate system; and determining the contact point based on the three-dimensional posture model and the three-dimensional environment model in the same coordinate system.

[0033] Optionally, unifying the robot's three-dimensional posture model and the target environment's three-dimensional environment model into the same coordinate system can ensure that when the robot processes and analyzes environmental information, its own posture data and the object position data in the environment are in the same reference frame, which is conducive to data fusion and processing, and avoids spatial positioning errors caused by inconsistent coordinate systems. Performing spatial calculations in a unified coordinate system can more accurately determine the contact points between the robot and environmental objects, which can include but are not limited to the contact points between the robot's operating tool and the target object, the contact points between the robot's chassis and the supporting surface, etc., which is crucial for the stability and safety of the robot when performing tasks. Based on the three-dimensional posture model and three-dimensional environment model in a unified coordinate system, the robot's motion trajectory can be planned more accurately to ensure that it can effectively complete the task when it contacts the object, while maintaining balance and avoiding collision or damage. When the robot encounters environmental changes or adjusts its own posture during movement, it can quickly recalculate the contact points in a unified coordinate system and adjust the motion plan, thereby improving the robot's dynamic response capability and environmental adaptability.

[0034] Optionally, the 3D posture model and the 3D environment model may be unified into the same coordinate system in the following manners, but not limited to:

[0035] Step S1061, coordinate system alignment: First, determine the origin of the unified coordinate system. You can choose the center of mass of the robot or a fixed and distinctive point in the environment as the origin. Determine the positive direction of the X, Y, and Z axes. Usually, the Z axis is perpendicular to the ground, and the X and Y axes are parallel to the ground.

[0036] Step S1062, perform robot posture model conversion, specifically, use the robot's sensor data, such as the joint angle measured by the encoder, to convert the coordinates of each point in the robot posture model from the local joint coordinate system to the unified world coordinate system. Perform environmental model conversion, specifically, perform coordinate conversion on external perception data, including but not limited to laser radar point cloud data and camera image data. These data need to be converted from the sensor coordinate system to the unified world coordinate system.

[0037] Step S1063: By fusing multiple sensor data, such as integrating the perception results of the laser radar and the camera, the accuracy and reliability of environmental perception are improved. Calibration technology is used to correct the errors of the sensor data to ensure that the data in the unified coordinate system is accurate.

[0038] In an optional embodiment, when the contact points include multiple first target contact points between an operating tool on the robot and a target object in a target environment, and multiple second target contact points between a chassis of the robot and a corresponding support surface, the contact points are determined based on a three-dimensional posture model and a three-dimensional environment model, including: determining multiple first initial contact points, multiple second initial contact points, and distance information between the multiple first initial contact points and the operating tool respectively based on the three-dimensional posture model and the three-dimensional environment model, wherein the first initial contact point is the initial contact point between the operating tool and the target object, and the second initial contact point is the initial contact point between the chassis of the robot and the corresponding support surface; based on the distance information between the multiple first initial contact points and the operating tool respectively, screening out multiple first target contact points from the multiple first initial contact points, and screening out multiple second target contact points from the multiple second initial contact points, wherein the number of the multiple first target contact points is less than or equal to the number of the multiple first initial contact points, and the number of the multiple second target contact points is less than or equal to the number of the multiple second initial contact points.

[0039] Optionally, by screening multiple first initial contact points (contact points between the operating tool and the target object) and multiple second initial contact points (contact points between the robot chassis and the support surface), the robot can more accurately select those contact points that are most conducive to task execution and balance. This refined selection can improve the efficiency and safety of robot operation. Through the analysis of distance information, the robot is allowed to intelligently adjust its motion strategy according to the distance between the operating tool and the target object, ensuring that the operating tool can contact the target in the shortest path or the most labor-saving way while completing the task, reducing energy consumption and operation time. The process of screening out target contact points from multiple initial contact points is actually optimizing the selection of contact points to ensure that the contact points selected by the robot are the most conducive to task execution. This includes selecting those points that can provide maximum operational stability, as well as those points that are closest to the operating tool and most convenient to operate, thereby improving the success rate of the task. By simultaneously considering the contact points between the operating tool and the target object (first target contact points) and the contact points between the robot chassis and the support surface (second target contact points), the robot can more effectively coordinate its movement and balance. Especially in dynamic environments, this helps the robot to complete tasks efficiently while maintaining stability. By selecting fewer target contact points from multiple initial contact points, the complexity of the robot's motor brain design algorithm can be reduced, making the control process more feasible. It can also reduce the uncertainty of the robot in performing tasks and improve the robustness of the control system.

[0040] Optionally, the initial contact point between the operating tool and the target object, and the initial contact point between the chassis of the robot and the corresponding support surface can be obtained by, but not limited to, performing posture calculation on the robot.

[0041] In an optional embodiment, based on the distance information between the multiple first initial contact points and the operating tools respectively, multiple first target contact points are screened out from the multiple first initial contact points, and multiple second target contact points are screened out from the multiple second initial contact points, including: taking the points in the multiple first initial contact points whose distance information is less than a predetermined distance as the multiple first target contact points; determining the movement information of the chassis based on the distance information between the first initial contact points and the operating tools respectively, wherein the movement information includes the movement trajectory and the movement distance; and screening out the multiple second target contact points from the multiple second initial contact points based on the movement information.

[0042] Optionally, by setting a predetermined distance threshold, the robot is allowed to select those points closest to the operating tool from a number of first initial contact points as the first target contact point. This optimization strategy based on distance information helps to reduce the time and energy required for the movement of the operating tool and improve the efficiency of the robot in performing tasks. The movement information is set to include a movement trajectory and a movement distance, and the movement information is calculated based on the distance information between the first initial contact point and the operating tool. This step can ensure that when the robot approaches the target object, the movement of its chassis can not only efficiently reach the predetermined position, but also maintain its own balance, thereby avoiding unnecessary collisions or stability problems. After determining the movement trajectory of the operating tool, the robot can more accurately select the second target contact point, that is, the contact point between the robot chassis and the supporting surface. The selection of these points will directly affect the stability of the robot during movement, ensuring that the robot can maintain the best balance when completing the task and reduce the risk of falling. Through the above screening mechanism, the robot can dynamically adjust the motion strategy of its operating tool and chassis in a constantly changing environment to adapt to environmental changes and task requirements, and improve the robustness of task execution (i.e., stability and reliability in the face of uncertainty and change). Screening the first target contact point closest to the operating tool and optimizing the chassis contact point based on movement information can significantly shorten the time from the robot sensing the environment to completing the task, and improve its response speed to emergency or dynamic environments.

[0043] Optionally, when determining the target contact point (i.e., the first target contact point and the second target contact point), first, the distance information between the operating tool and the target object is determined based on the point cloud data collected by the external perception system (such as laser radar, camera). This includes calculating the shortest distance and the precise position of the contact point. From all possible initial contact points, the contact point that is closest to the operating tool and most conducive to task execution is selected as the first target contact point. The center of mass position of the robot in the current posture is calculated using the data of the inertial measurement unit (IMU) and internal sensors (such as joint encoders and force / torque sensors). The center of mass position is a key parameter for balance control. Based on the center of mass position, the load of the operating tool and the kinematic parameters of the robot structure, a dynamic analysis is performed to predict the change in the stability of the robot under different motion states. It may include, but is not limited to, calculating the torque of the robot in each degree of freedom and analyzing the influence of different movements on the center of mass position. Based on the distance information and the results of the balance analysis, the movement trajectory and speed of the robot chassis are planned to ensure that the robot can maintain balance when completing the operation. It may involve, but is not limited to, adjusting the position of the supporting legs, calculating the direction and speed of movement, and predicting the immediate impact of movement on the stability of the robot. During the movement of the robot chassis, continuously monitor the changes in the contact points between the operating tool and the target object, the internal state of the robot (such as joint angles, speeds and torques), and changes in the external environment. Based on real-time feedback, dynamically adjust the chassis movement plan to ensure the balance and stability of the robot during movement. Convert the chassis movement information (including movement trajectory, speed and direction) into control instructions and send them to the robot actuator to drive the robot chassis to move along the planned path. At the same time, monitor the robot movement and environmental changes to respond to possible unexpected situations, such as unstable ground or changes in the position of the target object.

[0044] Step S108, performing motion planning for the robot based on the contact points to obtain a motion planning result;

[0045] In the above step S108, motion planning is performed based on contact points, which can ensure that the robot can not only complete the task efficiently when moving, but also maintain its own stability to avoid falls or damages caused by mistakes or external interference, while reducing energy consumption and improving operational efficiency.

[0046] Step S110, controlling the movement of the robot based on the motion planning result.

[0047] In the above step S110, the robot movement is controlled based on the motion planning result, and the whole process is run in a dynamically updated environment, which means that the robot can adjust its motion planning and control strategy according to real-time sensor data, which helps the robot to deal with emergencies, such as the appearance of obstacles or sudden changes in its own state, thereby improving its safety and reliability in practical applications.

[0048] In an optional embodiment, the robot motion is controlled based on the motion planning results, including: constructing a static model of the robot based on predetermined motion parameters corresponding to the multiple degrees of freedom of the robot, wherein the predetermined motion parameters include at least the rotation angle range, output speed range, and torque range of the corresponding degrees of freedom; obtaining a posture model of the robot based on the static model and current motion parameters corresponding to the multiple degrees of freedom collected during the robot motion, wherein the current motion parameters include at least: the output torque, speed, and position of the corresponding degrees of freedom; and controlling the robot motion based on the posture model and the motion planning results.

[0049] Optionally, by building a static model of the robot based on predetermined motion parameters corresponding to multiple degrees of freedom (such as rotation angle range, output speed range, torque range), the robot is helped to understand the physical limitations and capability boundaries of its own motion. This provides a basis for subsequent dynamic posture analysis and control, ensuring that the robot motion planning is carried out within the physically feasible range. During the robot motion process, based on the combination of the current motion parameters (output torque, speed and position) with the static model, a dynamic model reflecting the real-time posture of the robot is allowed to be generated. The analysis of this real-time posture model is crucial to ensure the balance and stability of the robot in motion, especially in multi-joint robots, which can adjust the motion of each joint in time according to the dynamic posture model to avoid tipping over or unstable state. Controlling the robot motion based on the posture model and motion planning results can ensure that the control instructions are highly accurate and can effectively guide the robot to complete the preset tasks. This precise control helps to reduce the errors of the robot during execution and improve the efficiency and success rate of the operation. By analyzing the static model and the dynamic posture model, the robot can better manage its energy consumption, for example, adjusting the motion parameters to save energy while ensuring the completion of the task, or optimizing the motion path to reduce unnecessary torque output and improve overall energy efficiency. The real-time posture model can ensure that the robot can accurately understand its own state at any time. Even when encountering accidents or environmental changes, it can restore balance or change task execution strategies by dynamically adjusting motion parameters, thereby enhancing the robustness and safety of the system.

[0050] Optionally, the static model is constructed based on the predetermined motion parameters of each degree of freedom of the robot (such as the rotation angle range, output speed range, and torque range), which reflects the structure and capabilities of the robot when there is no external force. When constructing the static model, the physical characteristics of each degree of freedom can be recorded in detail, including but not limited to the joint type (rotation or translation), joint limit (maximum and minimum rotation angle), and output capacity (maximum speed, maximum torque) of the power source (motor, hydraulic, pneumatic). In addition, the static model can also include the mass distribution and inertia moment of each part of the robot, which is the basis for dynamic analysis and balance control.

[0051] Optionally, the posture model can be understood as a real-time posture model, which can be generated based on the static model and the current motion parameters (such as the output torque, speed and position of the degrees of freedom) collected in real time during the robot's motion, and it reflects the actual state of the robot at a certain moment. When generating a real-time posture model, it is necessary to update the motion state of each degree of freedom in real time, which may include but is not limited to its angle, speed and torque. High-precision sensors such as joint encoders, force / torque sensors and inertial measurement units (IMUs) can be used. The real-time posture model not only reflects the current posture of the robot, but also predicts the evolution of the robot's posture over time under given torque and speed.

[0052] Optionally, when controlling the robot's motion based on the real-time posture model and motion planning results, the actual motion of the robot is guided by solving the motion parameters of each degree of freedom of the robot when completing a specific task, including torque, speed, and angle. This process involves inverse dynamics calculations to determine the torque that needs to be applied to each joint of the robot to achieve the motion planning results. In addition, the robot's dynamic characteristics, such as inertia, friction, and gravity, can be further considered to generate control signals that conform to the laws of physics. The control strategy should include speed control and torque control to ensure that the robot can perform tasks at a stable speed and torque while avoiding excessive forces on the joints or the environment.

[0053] In an optional embodiment, the method further includes: during the movement of the robot, when changes in internal state information and / or environmental information are detected, updating the three-dimensional posture model and / or three-dimensional environment model based on the new internal state information and new environmental information.

[0054] Optionally, during the robot's motion, when changes are detected in internal state information (such as joint speed, position, torque, etc.) and / or environmental information (such as the appearance of obstacles, changes in the position of target objects, etc.), the three-dimensional posture model and / or three-dimensional environmental model can be updated immediately based on the new information. This dynamic response capability enables the robot to quickly adapt to environmental changes and improve its operational flexibility and safety in dynamic and unpredictable environments. Once the three-dimensional model is updated, the robot can immediately adjust its motion planning to ensure that the motion trajectory and operation strategy can reflect the latest environmental conditions and its own state, thereby avoiding potential collisions or stability problems and improving the accuracy of task completion. Real-time model updating helps the robot to more accurately judge its relative position with environmental objects, adjust the motion strategy in time, and ensure that all operations are performed in the best state and position during the execution of the task, thereby improving the success rate of task completion. In addition, dynamic planning based on the latest state and environmental information can avoid unnecessary movements and operations, reduce energy consumption, and improve the economy and efficiency of robot operation.

[0055] Through the above steps S102 to S110, the purpose of comprehensive three-dimensional modeling and robot motion planning based on internal state information and environmental information can be achieved, thereby realizing the technical effect of improving the accuracy of robot path planning and operation balance, and further solving the technical problem in related technologies that factors are not fully considered when performing robot motion control, resulting in inaccurate path planning and dexterous operation, and poor balance of the robot during movement.

[0056] Based on the above embodiments and optional embodiments, the present invention proposes an optional implementation mode: Figure 2 is a flow chart of an optional robot motion brain design method according to an embodiment of the present invention, such as Figure 2 As shown, the method includes:

[0057] S1, through the robot's external perception system, perceives the robot's surrounding environment information, measures the distance from all objects around the robot to the robot, identifies and locates the target object, and obtains the environmental information of the robot's target environment.

[0058] S2, through the robot's internal perception system, perceives the speed, position and force information of all the robot's joints, evaluates the robot's center of mass position, performs preliminary identification of the robot's real-time posture, and obtains the robot's internal state information.

[0059] Based on the environmental information and the internal state information of the robot, the target environment in which the robot is located and the robot are respectively three-dimensionally modeled to obtain a three-dimensional environmental model of the target environment and a three-dimensional posture model of the robot.

[0060] S3, unify the robot's 3D posture model and the target environment's 3D environment model in a world coordinate system, specifically:

[0061] The 3D model in the world coordinate system is point cloud fused according to the robot's external sensor system, such as laser radar, camera, ultrasonic sensor, etc. The key to unifying the robot's 3D posture model with the 3D environment model is to align the origins of the two coordinate systems. There are two methods. The first is to use the robot's own center of mass as the origin, and the second is to use a certain point in the world environment as the origin.

[0062] S4, based on the 3D posture model and 3D environment model in the world coordinate system, perform spatial calculations on the robot and surrounding objects to determine the contact points between the robot and the target environment (i.e., the surrounding environment), wherein the contact points may include but are not limited to the initial contact points between the robot's operating tool and the target object, and the initial contact points between the robot chassis and the support surface. Specifically:

[0063] S41, determining whether the robot has a contact point with an object in the target environment (including a wall, a ground, a table, a target object, etc.).

[0064] S42, determining the distance depth information (ie, distance information) from all points of the target object (ie, the initial contact points between the target object and the operating tool) to the robot operating tool (gripper, robot hand, tooling, etc.).

[0065] S43, determining the trajectory and distance that the robot chassis (legged, wheeled, tracked, or other) needs to move under the premise that the robot operating tool can complete the operating task of the operated object.

[0066] S44, the robot relies on whole-body posture solution to determine the initial contact point between the robot operating tool and the target object, and the initial contact point between the robot chassis and the supporting surface.

[0067] S5, selecting the contact points between the robot and all target objects, including but not limited to the hands, feet, head and other points, and planning the motion trajectory of the contact points.

[0068] Specifically, based on the contact points obtained in step S4 above, the best contact points between the robot and the target object (including the ground, wall, desktop, tool to be grasped or operated, etc.) are further selected, and the best moving trajectory of the robot contact points is selected with the goal of achieving effective contact between the robot and several target objects, to obtain the robot's motion planning result. Among them:

[0069] The process of determining the best contact point (i.e., the first target contact point) is as follows: determine the set of contact points that can match the local three-dimensional features of the target object and the robot operation tool. In the above set, determine the contact point with the shortest distance from the target object to the robot operation tool. During the robot movement operation, the robot's motion trajectory is continuously corrected so that it can successfully complete the operation task.

[0070] The specific process of determining the effective contact point (i.e., the second target contact point) is as follows: for the robot chassis and the supporting surface, it means that the supporting surface can maintain the balance and stability of the robot; for the robot operating tool and the target object, it means that the local features of the target object match the local features of the robot operating tool.

[0071] The robot needs to maintain static balance or dynamic balance. If the robot is in static balance, it can contact any object. If the robot is in dynamic balance, regardless of whether it contacts other objects, the robot chassis needs to have a tendency to quickly catch up with the center of gravity.

[0072] S6, combined with the robot's own dynamics, calculates the posture of each joint of the robot, while taking into account the robot's own balance control. Specifically includes:

[0073] S61, establishing a static model of the robot, including the distance and direction of the robot's degree of freedom to the adjacent degree of freedom, and determining the rotation angle range and output speed and torque range of each degree of freedom;

[0074] S62, detecting the output torque and speed of each degree of freedom through a sensor to determine the posture model of the robot;

[0075] S63, determining the center of mass position of the robot in combination with the inertial sensor of the robot;

[0076] S64, based on inverse dynamics, calculates the dynamic changes of each degree of freedom of the robot after completing a certain movement.

[0077] S5, the robot executes the planning instruction and dynamically updates the world coordinates and its own coordinates, and repeats the operations of steps S1 to S6 until the robot completes the target task.

[0078] It should be noted that, in this embodiment, the perception system of the robot is divided into two parts: the external perception system (brain) and the internal perception system (cerebellum), which perceive the external environment information and the robot's internal posture information respectively. The three-dimensional size of the external environment and the robot's own posture information are simultaneously used as the necessary conditions for the robot's perception, so that the robot can efficiently find the target object and select the contact point. With any change in the information of the external environment or the robot's internal posture as the trigger condition, the robot's own posture change or the environment change will re-model the robot's posture in the world coordinate system, and high dynamics will maintain balance and stability, which will help the robot to quickly update the model, maintain balance and stability, and improve the speed of completing tasks. Taking the contact points between the robot and multiple target objects as the optimization target, selecting the best contact points and contact angles, and planning the contact point trajectory can not only maintain balance and stability, but also complete the operation task as quickly as possible.

[0079] In this embodiment, a robot is also provided, including: a robot body, on which a first sensor system and a second sensor system are arranged; and a controller, connected to the robot body, for executing any one of the above-mentioned robot control robot motion brain design methods.

[0080] In this embodiment, a robot motion brain design device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the terms "module" and "device" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0081] According to an embodiment of the present invention, there is also provided an embodiment of a device for implementing the above-mentioned robot motion brain design method. Figure 3 A schematic diagram of the structure of a robot motion brain design device according to an embodiment of the present invention is shown in FIG. Figure 3 As shown, the above robot motion brain design device includes: an information acquisition module 300, a three-dimensional modeling module 302, a contact point determination module 304, a motion planning module 306, and a motion control module 308, wherein:

[0082] An information collection module 300 is used to collect internal state information of the robot based on a first sensor system on the robot, and to collect environmental information of a target environment where the robot is located based on a second sensor system on the robot;

[0083] The three-dimensional modeling module 302 is connected to the information acquisition module 300 and is used to perform three-dimensional modeling on the robot and the target environment respectively based on the internal state information and the environmental information, so as to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment;

[0084] A contact point determination module 304, connected to the 3D modeling module 302, for determining the contact point between the robot and the target environment based on the 3D posture model and the 3D environment model;

[0085] The motion planning module 306 is connected to the contact point determination module 304 and is used to perform motion planning for the robot based on the contact points to obtain a motion planning result;

[0086] The motion control module 308 is connected to the motion planning module 306 and is used to control the motion of the robot based on the motion planning result.

[0087] In the embodiment of the present invention, an information acquisition module 300 is provided to collect the internal state information of the robot based on the first sensor system on the robot, and to collect the environmental information of the target environment of the robot based on the second sensor system on the robot; a three-dimensional modeling module 302 is connected to the information acquisition module 300, and is used to perform three-dimensional modeling of the robot and the target environment respectively based on the internal state information and the environmental information, so as to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment; a contact point determination module 304 is connected to the three-dimensional modeling module 302, and is used to determine the contact point between the robot and the target environment based on the three-dimensional posture model and the three-dimensional environment model. contact point; a motion planning module 306, connected to the contact point determination module 304, for performing motion planning for the robot based on the contact point to obtain a motion planning result; a motion control module 308, connected to the motion planning module 306, for controlling the robot motion based on the motion planning result, thereby achieving the purpose of comprehensive three-dimensional modeling and robot motion planning based on internal state information and environmental information, thereby achieving the technical effect of improving the accuracy of robot path planning and operation balance, and further solving the technical problem in the related art that factors are not fully considered when performing robot motion control, resulting in inaccurate path planning and dexterous operation, and poor balance of the robot during movement.

[0088] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0089] It should be noted that the above information collection module 300, 3D modeling module 302, contact point determination module 304, motion planning module 306, and motion control module 308 correspond to steps S102 to S110 in the embodiment. The examples and application scenarios implemented by the modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment. It should be noted that the above modules can be run in a computer terminal as part of the device.

[0090] It should be noted that the optional or preferred implementation of this embodiment can refer to the relevant description in the embodiment, which will not be repeated here.

[0091] The above-mentioned robot motion brain design device may also include a processor and a memory. The above-mentioned information acquisition module 300, three-dimensional modeling module 302, contact point determination module 304, motion planning module 306, motion control module 308, etc. are all stored in the memory as program modules, and the processor executes the above-mentioned program modules stored in the memory to realize corresponding functions.

[0092] According to an embodiment of the present application, a robot embodiment is also provided. The robot comprises: a robot body, on which a first sensor system and a second sensor system are arranged; and a controller connected to the robot body, for executing any one of the above-mentioned robot motion brain design methods.

[0093] The processor includes a kernel, which retrieves the corresponding program module from the memory. The kernel may be one or more. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. The memory includes at least one memory chip.

[0094] According to an embodiment of the present application, an embodiment of a non-volatile storage medium is also provided. Optionally, in this embodiment, the non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute any one of the above-mentioned robot motion brain design methods.

[0095] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group, and the non-volatile storage medium includes a stored program.

[0096] Optionally, when the program is running, the device where the non-volatile storage medium is located is controlled to perform the following functions: based on the first sensor system on the robot, the internal state information of the robot is collected; based on the second sensor system on the robot, the environmental information of the target environment in which the robot is located is collected; based on the internal state information and the environmental information, the robot and the target environment are respectively three-dimensionally modeled to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment; based on the three-dimensional posture model and the three-dimensional environment model, the contact points between the robot and the target environment are determined; based on the contact points, the robot is motion planned to obtain motion planning results; and the robot movement is controlled based on the motion planning results.

[0097] According to an embodiment of the present application, an embodiment of a processor is also provided. Optionally, in this embodiment, the processor is used to run a program, wherein the program executes any one of the above-mentioned robot motion brain design methods when it is run.

[0098] According to an embodiment of the present application, an embodiment of a computer program product is also provided, which, when executed on a data processing device, is suitable for executing a program that initializes any of the steps of the above-mentioned robot motion brain design method.

[0099] Optionally, the above-mentioned computer program product, when executed on a data processing device, is suitable for executing a program that is initialized with the following method steps: based on the first sensor system on the robot, the internal state information of the robot is collected, and based on the second sensor system on the robot, the environmental information of the target environment in which the robot is located is collected; based on the internal state information and the environmental information, three-dimensional modeling of the robot and the target environment is performed respectively to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment; based on the three-dimensional posture model and the three-dimensional environment model, the contact point between the robot and the target environment is determined; based on the contact point, motion planning is performed on the robot to obtain a motion planning result; and the movement of the robot is controlled based on the motion planning result.

[0100] An embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are implemented: based on a first sensor system on the robot, internal state information of the robot is collected; based on a second sensor system on the robot, environmental information of a target environment in which the robot is located is collected; based on the internal state information and environmental information, three-dimensional modeling is performed on the robot and the target environment respectively to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment; based on the three-dimensional posture model and the three-dimensional environment model, the contact point between the robot and the target environment is determined; based on the contact point, motion planning is performed on the robot to obtain a motion planning result; and the movement of the robot is controlled based on the motion planning result.

[0101] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0102] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0103] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the above modules can be a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, modules or indirect coupling or communication connection of modules, which can be electrical or other forms.

[0104] The modules described above as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0105] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0106] If the above-mentioned integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable non-volatile storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a non-volatile storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned non-volatile storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0107] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for designing a robot motion brain, characterized in that: include: Collecting internal state information of the robot based on a first sensor system on the robot, and collecting environmental information of a target environment where the robot is located based on a second sensor system on the robot; Based on the internal state information and the environmental information, three-dimensional modeling is performed on the robot and the target environment respectively to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment; Determining a contact point between the robot and the target environment based on the three-dimensional posture model and the three-dimensional environment model; Performing motion planning on the robot based on the contact points to obtain a motion planning result; The robot movement is controlled based on the motion planning result.

2. The method according to claim 1, characterized in that The step of determining the contact point between the robot and the target environment based on the three-dimensional posture model and the three-dimensional environment model includes: Unifying the three-dimensional posture model and the three-dimensional environment model into the same coordinate system; The contact point is determined based on the three-dimensional posture model and the three-dimensional environment model in the same coordinate system.

3. The method according to claim 1, characterized in that In a case where the contact points include a plurality of first target contact points between an operating tool on the robot and a target object in the target environment, and a plurality of second target contact points between a chassis of the robot and a corresponding support surface, determining the contact points based on the three-dimensional posture model and the three-dimensional environment model includes: Based on the three-dimensional posture model and the three-dimensional environment model, determine a plurality of first initial contact points, a plurality of second initial contact points, and distance information between the plurality of first initial contact points and the operating tool, wherein the first initial contact point is an initial contact point between the operating tool and the target object, and the second initial contact point is an initial contact point between the chassis of the robot and the corresponding support surface; Based on the distance information between the multiple first initial contact points and the operating tool respectively, the multiple first target contact points are screened out from the multiple first initial contact points, and the multiple second target contact points are screened out from the multiple second initial contact points, wherein the number of the multiple first target contact points is less than or equal to the number of the multiple first initial contact points, and the number of the multiple second target contact points is less than or equal to the number of the multiple second initial contact points.

4. The method according to claim 3, characterized in that The step of selecting the plurality of first target contact points from the plurality of first initial contact points based on the distance information between the plurality of first initial contact points and the operating tool, and selecting the plurality of second target contact points from the plurality of second initial contact points, comprises: Using points whose distance information is less than a predetermined distance among the plurality of first initial contact points as the plurality of first target contact points; Determine the movement information of the chassis based on the distance information between the first initial contact point and the operating tool, wherein the movement information includes a movement track and a movement distance; Based on the movement information, the plurality of second target contact points are screened out from the plurality of second initial contact points.

5. The method according to claim 1, characterized in that The three-dimensional modeling of the robot and the target environment is performed based on the internal state information and the environmental information to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment, including: During the movement of the robot, three-dimensional modeling is performed on the robot and the target environment respectively based on the internal state information and the environmental information to obtain the three-dimensional posture model and the three-dimensional environmental model.

6. The method according to claim 1, characterized in that The method further comprises: During the movement of the robot, when it is detected that the internal state information and / or the environmental information have changed, the three-dimensional posture model and / or the three-dimensional environmental model is updated based on the new internal state information and the new environmental information.

7. The method according to any one of claims 1 to 6, characterized in that: The controlling the robot movement based on the motion planning result comprises: Based on predetermined motion parameters corresponding to the multiple degrees of freedom of the robot, respectively, a static model of the robot is constructed, wherein the predetermined motion parameters at least include a rotation angle range, an output speed range, and a torque range of the corresponding degrees of freedom; Based on the static model and the current motion parameters corresponding to the multiple degrees of freedom respectively collected during the movement of the robot, the posture model of the robot is obtained, wherein the current motion parameters at least include: output torque, speed and position of the corresponding degrees of freedom; The robot movement is controlled based on the posture model and the motion planning result.

8. A robot motion brain design device, characterized in that: include: an information collection module, configured to collect internal state information of the robot based on a first sensor system on the robot, and collect environmental information of a target environment where the robot is located based on a second sensor system on the robot; A three-dimensional modeling module, configured to perform three-dimensional modeling on the robot and the target environment respectively based on the internal state information and the environmental information, so as to obtain a three-dimensional posture model corresponding to the robot and a three-dimensional environment model corresponding to the target environment; a contact point determination module, configured to determine a contact point between the robot and the target environment based on the three-dimensional posture model and the three-dimensional environment model; A motion planning module, used for performing motion planning for the robot based on the contact points to obtain a motion planning result; A motion control module is used to control the motion of the robot based on the motion planning result.

9. A robot, characterized in that: include: A robot body, wherein the robot body is provided with a first sensor system and a second sensor system; A controller, connected to the robot body, is used to execute the robot motion brain design method described in any one of claims 1 to 7.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the robot motion brain design method described in any one of claims 1 to 7.

Images