Robot multi-dimensional simulation platform and construction method thereof
Through the robot multi-dimensional simulation platform, combined with finite element analysis and machine learning algorithms, the multi-dimensional behavior of robots in complex environments is simulated, and the problem of difficulty in capturing the interaction of dynamics, structures and control strategies in the existing technology is solved, and efficient simulation is achieved Results presentation and data transmission improve the practicality and reliability of simulation results.
Patent Information
- Application Number
- CN202411235552.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-12-06
AI Technical Summary
Existing robot simulation technologies are mainly concentrated in a single dimension, and it is difficult to fully reflect the comprehensive performance of robots in complex and changing environments, and it is impossible to effectively capture the interaction and impact of dynamics, structural design, control strategies and multi-sensor information.
Design a robot multi-dimensional simulation platform, including model library module, sensor library module, environment simulation module, solution module and display module. Through the combination of finite element analysis and machine learning algorithms, the behavior and reaction of the robot in different environments will be simulated, and the use of The Socket protocol performs efficient data transfer.
The multi-dimensional simulation of robots is realized, and the structural stress and strain during the robot movement can be displayed in real time, which improves the practicality and reliability of simulation results, and improves the accuracy of robot design and development through multi-dimensional simulation.
Smart Images

Figure CN119089749A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a robot multi-dimensional simulation platform and a construction method thereof, belonging to the technical field of robot simulation. Background Art
[0002] The importance of robot simulation research is becoming increasingly prominent, becoming a key bridge between theoretical design and practical application. At present, robot simulation technology mainly focuses on simulation in specific dimensions. For example, dynamic simulation is used to analyze the force and motion trajectory of the robot during movement; structural simulation focuses on the rationality verification and strength analysis of the robot's structural design; and control simulation focuses on optimizing the control algorithm to ensure that the robot can accurately perform the predetermined task. Although these single-dimensional simulation methods have achieved remarkable results in their respective fields, their limitations are gradually emerging in the face of increasingly complex and changeable robot application scenarios.
[0003] Single-dimensional simulation often fails to fully reflect the comprehensive performance of robots in real environments. For example, in a dynamically changing operating environment, robots not only need to have stable dynamic characteristics, but also need to consider the robustness of structural design, the flexibility of control strategies, and the fusion processing of multi-sensor information. These factors are intertwined and jointly affect the overall performance of the robot, and single-dimensional simulation is difficult to capture this multi-dimensional interaction and influence.
[0004] Therefore, in order to more accurately simulate the behavioral characteristics of robots in complex and changing environments and improve the practicality and reliability of simulation results, it is necessary to integrate robot simulation technology from multiple dimensions. Summary of the invention
[0005] In order to solve the problems existing in the background technology, the present invention provides a robot multi-dimensional simulation platform and a construction method thereof.
[0006] To achieve the above purpose, the present invention adopts the following technical scheme: a robot multi-dimensional simulation platform, comprising
[0007] Model library module: built based on simulation requirements of different robot types and scenarios;
[0008] Sensor library module: including analog displacement sensors and visual sensors;
[0009] Environmental simulation module: simulates the light intensity and direction of different time periods by simulating lighting conditions;
[0010] Solution module: Based on finite element analysis and rigid body dynamics, the robot motion state is solved by introducing machine learning algorithms;
[0011] Display module: Map the solution results and robot motion process to the robot's three-dimensional model.
[0012] A method for constructing a robot multi-dimensional simulation platform, the method comprising the following steps:
[0013] S1: Configure the robot model library;
[0014] S101: Configuring robot actuators to simulate the operation and interaction of the robot in the real world;
[0015] S102: Configuring a robot walking mechanism to simulate the robot's walking and movement capabilities in different terrains and environments.
[0016] S2: Motion process simulation: Kinematic simulation and analysis of the motion of an object or system, providing intuitive visualization results;
[0017] S201: Through the motion simulation function of a single model, a normally distributed noise signal is added to the motion path of the model, so that the model exhibits random fluctuations in each simulation.
[0018] S202: Connecting various modular components through simulated assembly to simulate the effect and performance of the overall linkage of the robot;
[0019] S203: Simulate collision function.
[0020] S3: Create virtual environment: Create a variety of different environmental scenes and simulate the light intensity and direction of different time periods by simulating lighting conditions to evaluate the performance and adaptability of sensors under different environmental conditions;
[0021] S301: Testing the robot's performance in various environments;
[0022] S302: Test the performance of the binocular sensor at different times, analyze the dynamic response capability of the sensor at different time periods, and its application in target detection and scene change detection;
[0023] S303: Test the robot's adaptability on different terrains.
[0024] S4: Simulate the working principle and characteristics of each sensor, obtain the environmental information perceived by the sensor in real time, and perform data analysis and processing;
[0025] The simulation of the sensor includes the simulation of the binocular camera, the simulation of the laser radar and the simulation of the displacement sensor.
[0026] S5: Use Socket protocol for data transmission;
[0027] The method of using the Socket protocol to transfer data includes:
[0028] The communication unit within the host enables the information within the host to support each other and form a collaborative system;
[0029] Multi-host communication unit, can be combined with all software supporting Socket protocol.
[0030] S6: Use finite element analysis to build a model, solve the global stress and strain, predict the deformation results of the robot under various force conditions, and accelerate the solution process by establishing a proxy model.
[0031] The prediction method of the deformation result of the robot under various force conditions includes solving rigid body dynamics and establishing a lightweight proxy model through a machine learning algorithm using existing finite element analysis data related to the robot as a training set.
[0032] Compared with the prior art, the present invention has the following beneficial effects:
[0033] 1. The present invention combines multi-dimensional robot simulation technology to simulate the behaviors and reactions of various robots in different environments, thus realizing multi-dimensional simulation of robots;
[0034] 2. The present invention uses a finite element analysis proxy model to solve the robot structural deformation and displays it through a three-dimensional model, thereby realizing real-time stress and strain analysis of the robot structure during the robot movement;
[0035] 3. The communication module of the present invention adopts the Socket protocol for data transmission, thereby realizing efficient communication between the communication unit within the host and the communication units of multiple hosts. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 It is a structural block diagram of the robot multi-dimensional simulation platform of the present invention;
[0037] Figure 2 It is a schematic diagram of a method for constructing a robot multi-dimensional simulation platform of the present invention;
[0038] Figure 3 It is a schematic diagram for configuring the robot model library;
[0039] Figure 4 It is a schematic diagram of the motion process simulation;
[0040] Figure 5 It is a schematic diagram for creating a virtual environment;
[0041] Figure 6 It is a schematic diagram simulating the working principle and characteristics of each sensor;
[0042] Figure 7 This is a schematic diagram of using the Socket protocol for data transmission;
[0043] Figure 8 It is a schematic diagram for predicting the deformation results of the robot under various force conditions;
[0044] Fig. 9 It is a schematic diagram of the working method of the robot multi-dimensional simulation platform of the present invention. DETAILED DESCRIPTION
[0045] The technical solution 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 invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0046] A robot multi-dimensional simulation platform, comprising
[0047] Model library module: built based on simulation requirements of different robot types and scenarios;
[0048] Sensor library module: including analog displacement sensors and visual sensors;
[0049] Environmental simulation module: simulates the light intensity and direction of different time periods by simulating lighting conditions;
[0050] Solution module: Based on finite element analysis and rigid body dynamics, the robot motion state is solved by introducing machine learning algorithms;
[0051] Display module: Map the solution results and robot motion process to the robot's three-dimensional model.
[0052] A method for constructing a robot multi-dimensional simulation platform, the method comprising the following steps:
[0053] S1: Configure the robot model library to expand and customize the simulation platform as needed to meet the simulation requirements for different robot types and scenarios;
[0054] S101: Configure robotic actuators, such as mechanical arms and electric cylinders, with highly accurate motion control and force feedback systems to simulate the operation and interaction of robots in the real world;
[0055] S102: Configuring robot walking mechanisms, such as a walking chassis, which have high flexibility and stability and are used to simulate the robot's walking and movement capabilities in different terrains and environments.
[0056] S2: Motion process simulation: Kinematic simulation and analysis of the motion of an object or system. Using mathematical models and calculation methods, the simulation platform can simulate the robot's motion trajectory, speed, acceleration and other parameters under different conditions, and provide intuitive visualization results in combination with the robot's motion process;
[0057] S201: Through the precise motion simulation function of a single model, a normally distributed noise signal is added to the model's motion path, so that the model exhibits small random fluctuations in each simulation. These random fluctuations may affect the model's trajectory, speed or posture, making the model's motion look more realistic and vivid;
[0058] S202: Connecting various modular components through simulated assembly to simulate the effect and performance of the overall linkage of the robot;
[0059] This project will perform a series of logic to determine the validity of the connection, including checking the connection point matching and physical conflict. It also ensures that the assembled robot can operate stably through basic physical simulations, such as gravity and collision detection. At the same time, the intuitive user interface and real-time feedback mechanism enable users to easily drag and adjust the position and angle of parts, view the preview effect of the robot, and make necessary modifications or tests.
[0060] S203: A collision simulation function determines the contact between models through collision detection, and then performs a collision response involving changes in position, velocity or force to simulate the interaction of objects.
[0061] S3: Create virtual environment: Create a variety of different environmental scenes as needed, and simulate the light intensity and direction of different time periods by simulating lighting conditions, so as to more accurately evaluate the performance and adaptability of sensors under different environmental conditions;
[0062] S301: Testing the performance of the robot in various environments, for example, switching between indoor and outdoor scenes, simulating the robot's working conditions outdoors and indoors;
[0063] S302: Test the performance of the binocular sensor at different times. Adjust the light source position, light intensity and light source angle through simulation to simulate the impact of different weather conditions (such as sunny days, cloudy days, and nights) on the sensor performance. Study the differences in sensor response speed and sensitivity in different time periods (such as day and night). Analyze the dynamic response capability of the sensor in different time periods, especially in the application of target detection and scene change detection.
[0064] S303: Test the robot's adaptability on different terrains. For example, generate random obstacles and examine the performance of the perception algorithm used by the robot when facing different types of obstacles, including the accuracy and speed of obstacle detection.
[0065] S4: Accurately simulate the working principle and characteristics of each sensor, obtain the environmental information perceived by the sensor in real time, and perform accurate data analysis and processing;
[0066] Sensor simulation includes
[0067] Simulate the binocular camera and obtain binocular image data in real scenes to provide more realistic visual information for robot perception, positioning, navigation and other tasks. Perform the binocular sensor calibration process in the simulation environment to ensure that the image data of the left and right cameras can be accurately matched and aligned.
[0068] Laser radar simulation to detect possible obstacles, determine whether there are obstacles by processing laser detection results, and obtain the location of the collision point. In order to simulate the scanning process of laser radar, the ray detection function LineTraceByChannel is used to simulate the laser radar during the construction of the simulation platform, and the ray detection is repeated in a loop or iteration according to the scanning range and resolution of the laser radar;
[0069] Displacement sensor simulation, by simulating the behavior and output of the displacement sensor, can evaluate the robot's motion control performance in real applications.
[0070] S5: Supports the use of Socket protocol for data transmission, achieving efficient communication between the simulation platform and external programs, and providing a more flexible and efficient simulation solution;
[0071] The methods of using Socket protocol for data transmission include:
[0072] The communication unit within the host enables the information within the host to support each other and form a collaborative system;
[0073] Multi-host communication unit, can be combined with all software supporting Socket protocol.
[0074] This communication can be point-to-point or many-to-many, depending on the actual application requirements and network topology. In the real-time control process, the requirements for data transmission include low latency and no blocking. Therefore, the UDP protocol is selected in the Socket protocol for data transmission. The characteristics of UDP enable it to send data packets quickly without establishing a long connection, thereby reducing communication delays and providing efficient data transmission when the network conditions are good.
[0075] S6: Use finite element analysis to build a model, solve the global stress and strain, predict the deformation results of the robot under various force conditions, and accelerate the solution process by establishing a proxy model.
[0076] The prediction methods of the robot's deformation results under various force conditions include:
[0077] Solve rigid body dynamics, apply Newton's laws and angular momentum theorem and other basic principles to establish motion equations, describe the change of the robot's position over time during the movement process and the forces on key nodes;
[0078] Through machine learning algorithms, the existing finite element analysis data related to the robot is used as a training set to establish a lightweight proxy model. The lightweight proxy model inputs the forces of key nodes to simulate the stress and deformation of the robot under different force conditions.
[0079] The present invention has the ability to simulate various types of sensors such as binocular cameras, displacement sensors, and laser radars, and has a built-in modular robot component library including various types of actuators and walking mechanisms. It can accurately simulate component deformation through real-time finite element technology, and also provides a high-fidelity, interactive virtual environment that can be used to simulate and test various functions and behaviors of robots. On this basis, according to the specific needs of the robot, the simulation platform can assist users in selecting and matching modular components, and verify and optimize the robot configuration and control strategy by performing corresponding tasks, thereby guiding the design and development of the robot. In addition, the simulation platform takes into account the characteristics of simple operation, rich functions, and high simulation integration. It can stimulate the learning enthusiasm of robot enthusiasts with the help of visual interaction and target drive, and further promote the popularization of robot technology.
[0080] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations within the meaning and range of equivalents of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
[0081] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.
Claims
1. A robot multi-dimensional simulation platform, characterized in that: include Model library module: built based on simulation requirements of different robot types and scenarios; Sensor library module: including analog displacement sensors and visual sensors; Environmental simulation module: simulates the light intensity and direction of different time periods by simulating lighting conditions; Solution module: Based on finite element analysis and rigid body dynamics, the robot motion state is solved by introducing machine learning algorithms; Display module: Map the solution results and robot motion process to the robot's three-dimensional model.
2. A method for constructing a robot multi-dimensional simulation platform according to claim 1, characterized in that: The method comprises the following steps: S1: Configure the robot model library; S2: Motion process simulation: Kinematic simulation and analysis of the motion of an object or system, providing intuitive visualization results; S3: Create virtual environment: Create a variety of different environmental scenes and simulate the light intensity and direction of different time periods by simulating lighting conditions to evaluate the performance and adaptability of sensors under different environmental conditions; S4: Simulate the working principle and characteristics of each sensor, obtain the environmental information perceived by the sensor in real time, and perform data analysis and processing; S5: Use Socket protocol for data transmission; S6: Use finite element analysis to build a model, solve the global stress and strain, predict the deformation results of the robot under various force conditions, and accelerate the solution process by establishing a proxy model.
3. The method according to claim 2, characterized in that: The S1 comprises the following steps: S101: Configuring robot actuators to simulate the operation and interaction of the robot in the real world; S102: Configuring a robot walking mechanism to simulate the robot's walking and movement capabilities in different terrains and environments.
4. The method according to claim 2, characterized in that: The S2 comprises the following steps: S201: Through the motion simulation function of a single model, a normally distributed noise signal is added to the motion path of the model, so that the model exhibits random fluctuations in each simulation; S202: Connecting various modular components through simulated assembly to simulate the effect and performance of the overall linkage of the robot; S203: Collision simulation function.
5. The method according to claim 2, characterized in that: The S3 comprises the following steps: S301: Testing the robot's performance in various environments; S302: Test the performance of the binocular sensor at different times, analyze the dynamic response capability of the sensor at different time periods, and its application in target detection and scene change detection; S303: Test the robot's adaptability on different terrains.
6. The method according to claim 2, characterized in that: The simulation of the sensor in S4 includes the simulation of the binocular camera, the simulation of the laser radar and the simulation of the displacement sensor.
7. The method according to claim 2, characterized in that: The methods of using the Socket protocol for data transmission described in S5 include: The communication unit within the host enables the information within the host to support each other and form a collaborative system; Multi-host communication unit, can be combined with all software supporting Socket protocol.
8. The method according to claim 2, characterized in that: S6 The prediction method of the deformation results of the robot under various force conditions includes solving rigid body dynamics and using a machine learning algorithm to establish a lightweight proxy model using existing finite element analysis data related to the robot as a training set.
