IsaacSim-based humanoid robot simulation method for steel inspection and assay laboratory

Abstract

The invention discloses an IsaacSim-based humanoid robot simulation method for a steel inspection and test laboratory, and relates to the technical field of humanoid robot simulation methods, and the IsaacSim-based humanoid robot simulation method for the steel inspection and test laboratory comprises the following steps: step 1, carrying out refined modeling on environmental elements of the steel inspection and test laboratory; step 2, integrating a humanoid robot model in an IsaacSim simulation environment; 3, configuring workflows of different tasks of the robot in the simulation scene; 4, collecting and analyzing data when the robot executes the task in the virtual environment; and step 5, feeding back an adjustment algorithm or a control parameter based on a simulation result so as to improve the adaptability and the stability of the robot in a complex laboratory environment. Through the method, a user can perform comprehensive algorithm verification in a simulation environment, including a series of key algorithms such as a control algorithm, a visual identification algorithm and a motion planning algorithm of the humanoid robot, so that the stability and effectiveness of the humanoid robot in an actual environment are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of humanoid robot simulation methods, and in particular to an IsaacSim-based humanoid robot simulation method for a steel inspection and testing laboratory. Background Art

[0002] Robots can replace humans in steel inspection and testing laboratories to conduct experiments, avoiding unexpected risks that may occur during manual inspection. The application of robots in actual scenarios usually requires multiple verifications to ensure the possibility of conducting experiments in actual scenarios.

[0003] The defects of existing robot simulation methods are: 1. The prior art KR101475210B1 discloses APPRARATUS for simulating robots. However, this technology does not have a method for simulating key algorithms of humanoid robots in a virtual environment, and cannot provide feedback and improve the algorithms in a simulated environment. Therefore, a humanoid robot simulation method for a steel inspection and testing laboratory based on IsaacSim that can provide feedback and improve the simulated behavior of the robot in a simulated environment is needed to solve this problem.

[0004] 2. The prior art US5366896A discloses a robot-operated laboratory system, which does not provide technology for the robot to perform virtual simulation. When the robot conducts experiments in the laboratory, it cannot avoid the risks that may arise during the experiment through the virtual environment. When the robot simulation fails, it is easy to cause actual item loss. Therefore, a humanoid robot simulation method for steel inspection and testing laboratories based on IsaacSim is needed to solve this problem, which can avoid the potential risks that the robot may cause to itself and the surrounding environment through virtual simulation.

[0005] 3. The prior art KR1020130122288A discloses a robot three-dimensional simulation method and system. This technology does not provide a method for the robot to perform behavioral testing in a variety of simulated task environments. The robot cannot perform behavioral testing in a variety of scenarios. Therefore, a humanoid robot simulation method for a steel inspection and testing laboratory based on IsaacSim that can simulate robot behavior in a variety of different task environments is needed to solve this problem.

[0006] 4. The prior art CN115801869B discloses a patrol robot simulation system and method based on a distributed architecture. The robot debugging simulation cannot be completed through the simulation method. The robot accessories need to be debugged multiple times during actual debugging, which increases the debugging time. Therefore, a steel inspection and testing laboratory humanoid robot simulation method based on IsaacSim that can debug the robot accessories through a simulation method is needed to solve this problem. Summary of the invention

[0007] One purpose of the present application is to provide a humanoid robot simulation method for a steel inspection and testing laboratory based on IsaacSim, which can solve the technical problems raised in the prior art.

[0008] To achieve the above object, the present invention provides the following technical solution: a humanoid robot simulation method for a steel inspection and testing laboratory based on IsaacSim, the humanoid robot simulation method for a steel inspection and testing laboratory based on IsaacSim comprising the following steps: Step 1: Carry out detailed modeling of the environmental elements of the steel inspection laboratory, carry out detailed modeling of the key components and environmental elements of the steel inspection laboratory, establish a digital model consistent with the actual environment, and deploy the model to the database of the IsaacSim simulation environment to ensure that the simulation scene accurately reflects the operating environment and spatial layout of the real laboratory; Step 2: Integrate the humanoid robot model in the IsaacSim simulation environment and load the control system, including visual recognition, motion planning, and control algorithm modules, so that it can perform specific tasks in the simulation scenario. By setting perception sensors and motion drives for the robot model, the perception sensors include cameras and lidars, so that it has the same environmental perception and motion capabilities as a real robot. Step 3: Configure the workflow of different robot tasks in the simulation scenario. Set up the typical task flow of the humanoid robot in the simulation scenario, including sampling, analysis, classification and handling tasks in the inspection laboratory. By executing the task sequence in the simulation, obtain the performance data of the algorithm in the experimental task, verify and optimize the robot's control accuracy and task completion efficiency; Step 4: Collect and analyze data when the robot performs tasks in the virtual environment. Collect and analyze data on the performance of the humanoid robot in different task scenarios during the simulation process, including visual recognition accuracy, time efficiency of path planning, operational stability, collision detection parameters, and comprehensive evaluation of the algorithm performance, and generate an experimental report; Step 5: Adjust the algorithm or control parameters based on the simulation result feedback to improve the robot's adaptability and stability in complex laboratory environments.

[0009] Preferably, in step 1, a virtual environment of a steel inspection laboratory is created on an IsaacSim simulation platform, the environment including equipment and tools in the laboratory, and realistic physical interactions between the equipment in the virtual environment and the humanoid robot are ensured by configuring corresponding physical properties for the equipment and tools in the laboratory; Equipment layout: According to the equipment layout of the actual steel inspection and testing laboratory, the specific location of the equipment is set in the simulation environment to ensure that the simulation environment is consistent with the real laboratory; Detailed modeling of tools: Detailed modeling and functional configuration are performed for the tools that the robot needs to use in the experiment so that the robot can operate effectively in the simulation.

[0010] Preferably, the humanoid robot model in step 2 includes kinematic and dynamic models, sensor simulation and robot physical properties. The kinematic and dynamic models include its skeleton, joints, actuators and sensors, ensuring that the model imitates human joint movements and flexibly performs complex laboratory tasks of grasping, carrying and operating instruments. Each joint is simulated using a high-precision physics engine to ensure that the robot's behavior in the simulation environment is consistent with the actual hardware.

[0011] Robot Physics: Set the robot’s physical properties, including mass, center of gravity, and inertia, so that the robot exhibits realistic motion in the virtual environment, simulating the loads and dynamic behaviors of real work.

[0012] Preferably, the sensor simulates an integrated visual sensor and an IMU sensor to capture information in a virtual environment and feed it back to the control system to help the robot perform precise operations in the simulation.

[0013] Preferably, step 2 includes the construction of NVIDIA Omnigraph, task flow design and feedback control, and the construction of NVIDIA Omnigraph: Omnigraph is used for communication between the virtual environment and the external ROS system, and through Omnigraph, the robot is helped to interact with the external system in the virtual environment; Task flow design: The tasks of the humanoid robot in steel inspection and testing are designed by the robot control algorithm, including grabbing samples, pouring reagents, operating heating equipment, etc.; Feedback control: Through Omnigraph, Isaac Sim publishes sensor feedback data, which the robot can use to adjust its movements in real time, dynamically optimizing the robot's operations.

[0014] Preferably, the humanoid robot model in step 2 further includes ROS2 communication integration, two-way communication and control interface, ROS2 communication integration: simulating the interaction between the real robot and the external system, using ROS2 as the communication middleware between the robot and the simulation environment, and integrating it with IsaacSim to ensure the consistency between the simulation and the real system; Two-way communication: Through ROS2, the robot receives task instructions from the control system and feeds back its internal state and sensor data to the external system, completely simulating the behavior of the real robot communicating with the outside world during the simulation process; Control interface: ROS2 provides a standardized control interface for the robot, so that the motion control, data collection and feedback mechanisms in the simulation are consistent with the real robot. After the simulation is completed, the ROS2 communication part is applied to the actual robot deployment to reduce the debugging time in the actual environment.

[0015] Preferably, the step 3 designs the task flow of the robot in the steel inspection process based on the virtual environment and the robot model, including the following key operation steps: Sample collection: The robot grabs and takes out the steel sample from the sample container and places it in the designated testing equipment; Reagent addition: The robot accurately pours the required chemical reagents from the reagent bottle and adds them to the sample for chemical reaction; Equipment operation: The robot operates experimental equipment by simulating hand movements; Result reading: The robot reads the result data displayed by the experimental equipment through the built-in sensor, and transmits the data back to the control center for processing through the simulation software; These operations are simulated in real time using IsaacSim's physics engine to ensure the accuracy of the robot's movements and its adaptability to the laboratory environment.

[0016] Preferably, the different task scenarios in step 4 include normal working process and abnormal scenario testing, the normal working process: simulating the standard operating process in the steel inspection laboratory to ensure that the robot completes the whole process from sample processing to data reading; Abnormal scenario testing: When unexpected events are involved in the simulation, verify whether the robot can respond appropriately in unexpected situations and avoid further losses.

[0017] Preferably, the different task scenarios in step 4 also include extreme working condition testing: simulating complex or extreme environmental conditions to test the robot's operating performance and stability in these environments.

[0018] Preferably, in step 5, during the simulation operation, the system records each step of the robot operation, including the motion path, operation time, task success rate and equipment collision situation, and identifies the operation bottlenecks and potential risk points of the robot by analyzing these data. The specific analysis methods include: path optimization: analyzing the motion trajectory of the robot when performing the task, optimizing its joint movement and path planning, reducing unnecessary actions, and improving work efficiency; Collision detection: record the collision between the robot and the environment or equipment, adjust the action planning and operation accuracy, and avoid equipment damage or robot damage in actual operation; Iterative optimization is performed based on simulation data to adjust the robot's motion control algorithm and task flow to ensure that it can complete laboratory tasks safely and efficiently in a real laboratory environment.

[0019] Compared with the prior art, the present invention has the following beneficial effects: 1. Through this method, the present invention enables users to conduct comprehensive algorithm verification in a simulation environment, including a series of key algorithms such as the control algorithm, visual recognition algorithm, motion planning algorithm, etc. of the humanoid robot, to ensure its stability and effectiveness in the actual environment.

[0020] 2. The present invention effectively avoids the potential risks that the robot may cause to itself and its surrounding environment, and provides a safe and controllable testing platform, thereby optimizing the robot's operating efficiency and task completion quality, and promoting the widespread application of humanoid robots in inspection and testing scenarios in the steel and metallurgical industry.

[0021] 3. The present invention can test the operational performance and stability of the robot in these environments by simulating the performance of the humanoid robot in different task scenarios, including normal working procedures, abnormal scenario tests, and abnormal scenario tests, thereby reducing the simulation cost of the robot in actual scenarios.

[0022] 4. The present invention simplifies the behavior and control of the humanoid robot by simulating communication and feedback, thereby reducing the debugging time of the humanoid robot in the actual environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention is a flow chart of the humanoid robot simulation method for a steel inspection and testing laboratory. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0025] See also Figure 1 A humanoid robot simulation method for a steel inspection and testing laboratory based on IsaacSim, the humanoid robot simulation method for a steel inspection and testing laboratory based on IsaacSim comprises the following steps: Step 1: Carry out detailed modeling of the environmental elements of the steel inspection laboratory, carry out detailed modeling of the key components and environmental elements of the steel inspection laboratory, establish a digital model consistent with the actual environment, and deploy the model to the database of the IsaacSim simulation environment to ensure that the simulation scene accurately reflects the operating environment and spatial layout of the real laboratory; Step 2: Integrate the humanoid robot model in the IsaacSim simulation environment and load the control system, including visual recognition, motion planning, and control algorithm modules, so that it can perform specific tasks in the simulation scenario. By setting perception sensors and motion drives for the robot model, the perception sensors include cameras and lidars, so that it has the same environmental perception and motion capabilities as a real robot. Step 3: Configure the workflow of different robot tasks in the simulation scenario. Set up the typical task flow of the humanoid robot in the simulation scenario, including sampling, analysis, classification and handling tasks in the inspection laboratory. By executing the task sequence in the simulation, obtain the performance data of the algorithm in the experimental task, verify and optimize the robot's control accuracy and task completion efficiency; Step 4: Collect and analyze data when the robot performs tasks in the virtual environment. Collect and analyze data on the performance of the humanoid robot in different task scenarios during the simulation process, including visual recognition accuracy, time efficiency of path planning, operational stability, collision detection parameters, and comprehensive evaluation of the algorithm performance, and generate an experimental report; Step 5: Adjust the algorithm or control parameters based on the simulation result feedback to improve the robot's adaptability and stability in complex laboratory environments.

[0026] Step 1: Create a virtual environment of a steel inspection laboratory on the IsaacSim simulation platform. The environment includes the equipment and tools in the laboratory. By configuring the corresponding physical properties of the equipment and tools in the laboratory, ensure that the equipment in the virtual environment and the humanoid robot have realistic physical interactions. Equipment layout: According to the equipment layout of the actual steel inspection and testing laboratory, the specific location of the equipment is set in the simulation environment to ensure that the simulation environment is consistent with the real laboratory; Detailed modeling of tools: Detailed modeling and functional configuration are performed for the tools that the robot needs to use in the experiment so that the robot can operate effectively in the simulation.

[0027] The humanoid robot model in step 2 includes kinematic and dynamic models, sensor simulation and robot physical properties. The kinematic and dynamic models include its skeleton, joints, actuators and sensors to ensure that the model imitates human joint movements and can flexibly perform complex laboratory tasks such as grasping, carrying and operating instruments. Each joint is simulated using a high-precision physics engine to ensure that the robot's behavior in the simulation environment is consistent with the actual hardware.

[0028] Robot Physics: Set the robot’s physical properties, including mass, center of gravity, and inertia, so that the robot exhibits realistic motion in the virtual environment, simulating the loads and dynamic behaviors of real work.

[0029] Sensor simulation integrates visual sensors and IMU sensors to capture information in the virtual environment and feed it back to the control system to help the robot perform precise operations in simulation.

[0030] Step 2 includes the construction of NVIDIA Omnigraph, task flow design and feedback control. Build NVIDIA Omnigraph: Omnigraph is used for communication between the virtual environment and the external ROS environment. Through Omnigraph, the virtual robot can exchange data with external systems (such as ROS). Task flow design: The robot control algorithm is responsible for the logic of the robot's task execution, including grabbing samples, pouring reagents, and operating heating equipment. Omnigraph serves as a communication bridge to help the robot obtain task instructions and data from the external environment in the virtual environment and pass these instructions to the robot for execution; Feedback control: Through Omnigraph, Isaacsim publishes sensor information in the virtual environment to the external environment, including IMU information, image information and robot status information to the control algorithm. Combined with the feedback information from the sensor, the control algorithm adjusts the robot's movements in real time to ensure the continuity and accuracy of the movements, allowing the robot control algorithm to obtain specific environmental information and robot status information.

[0031] The humanoid robot model in step 2 also includes ROS2 communication integration, two-way communication and control interface. ROS2 communication integration: simulates the interaction between the real robot and the external system, uses ROS2 as the communication middleware between the robot and the simulation environment, and integrates it with IsaacSim to ensure the consistency between the simulation and the real system; Two-way communication: Through ROS2, the robot receives task instructions from the control system and feeds back its internal state and sensor data to the external system, completely simulating the behavior of the real robot communicating with the outside world during the simulation process; Control interface: ROS2 provides a standardized control interface for the robot, so that the motion control, data collection and feedback mechanisms in the simulation are consistent with the real robot. After the simulation is completed, the ROS2 communication part is applied to the actual robot deployment to reduce the debugging time in the actual environment.

[0032] Step 3: Based on the virtual environment and robot model, design the robot's task flow in the steel inspection process, including the following key operation steps: Sample collection: The robot grabs and takes out the steel sample from the sample container and places it in the designated testing equipment; Reagent addition: The robot accurately pours the required chemical reagents from the reagent bottle and adds them to the sample for chemical reaction; Equipment operation: The robot operates experimental equipment by simulating hand movements; Result reading: The robot reads the result data displayed by the experimental equipment through the built-in sensor, and transmits the data back to the control center for processing through the simulation software; These operations are simulated in real time using IsaacSim's physics engine to ensure the accuracy of the robot's movements and its adaptability to the laboratory environment.

[0033] The different task scenarios in step 4 include normal working process and abnormal scenario testing. Normal working process: simulate the standard operating process in the steel inspection and testing laboratory to ensure that the robot completes the entire process from sample processing to data reading; Abnormal scenario testing: When unexpected events are involved in the simulation, verify whether the robot can respond appropriately in unexpected situations and avoid further losses.

[0034] The different task scenarios in step 4 also include extreme working condition tests: simulating complex or extreme environmental conditions to test the robot's operating performance and stability in these environments.

[0035] In step 5, during the simulation operation, the system records each step of the robot operation, including the motion path, operation time, task success rate and equipment collision. By analyzing these data, the robot's operation bottlenecks and potential risk points are identified. The specific analysis methods include: Path optimization: analyzing the robot's motion trajectory when performing tasks, optimizing its joint motion and path planning, reducing unnecessary movements, and improving work efficiency; Collision detection: record the collision between the robot and the environment or equipment, adjust the action planning and operation accuracy, and avoid equipment damage or robot damage in actual operation; Iterative optimization is performed based on simulation data to adjust the robot's motion control algorithm and task flow to ensure that it can complete laboratory tasks safely and efficiently in a real laboratory environment.

[0036] See also Figure 1 , the present invention provides an embodiment; The simulation method of humanoid robot in steel inspection laboratory based on IsaacSim includes the following steps: Step 1: Carry out detailed modeling of the environmental elements of the steel inspection laboratory, carry out detailed modeling of the key components and environmental elements of the steel inspection laboratory, establish a digital model consistent with the actual environment, and deploy the model to the database of the IsaacSim simulation environment to ensure that the simulation scene accurately reflects the operating environment and spatial layout of the real laboratory; Create a virtual environment of a steel inspection laboratory on the IsaacSim simulation platform. The environment includes laboratory equipment and tools, such as test benches, chemical analysis instruments, sample processing equipment, heaters, cooling devices, etc. Configure the corresponding physical properties such as materials, friction coefficients, and gravity for the equipment and tools in the laboratory to ensure realistic physical interactions between the equipment and the humanoid robot in the virtual environment. Equipment layout: According to the equipment layout of the actual steel inspection and testing laboratory, the specific location of the equipment is set in the simulation environment to ensure that the simulation environment is consistent with the real laboratory; Detailed modeling of tools: Detailed modeling and functional configuration are performed for the tools that the robot needs to use in the experiment, such as samplers, beakers, stirrers, etc., so that the robot can operate effectively in the simulation.

[0037] Step 2: Integrate the humanoid robot model in the IsaacSim simulation environment and load the control system, including visual recognition, motion planning, and control algorithm modules, so that it can perform specific tasks in the simulation scenario. By setting perception sensors and motion drives for the robot model, the perception sensors include cameras and lidars, so that it has the same environmental perception and motion capabilities as a real robot. The humanoid robot model in step 2 includes kinematic and dynamic models, sensor simulation and robot physical characteristics. The kinematic and dynamic models include its skeleton, joints, actuators and sensors, etc., to ensure that the model imitates human joint movements and can flexibly perform complex laboratory tasks such as grasping, carrying and operating instruments. Each joint is simulated using a high-precision physics engine to ensure that the robot's behavior in the simulation environment is consistent with the actual hardware.

[0038] Robot physics: Set the robot's physical properties, including mass, center of gravity, and inertia, so that the robot can exhibit realistic motion in a virtual environment and simulate the loads and dynamic behaviors of actual work.

[0039] Sensor simulation integrates a variety of sensor modules, including visual sensors and IMU sensors, to capture information in the virtual environment and feed it back to the control system to help the robot perform precise operations in the simulation.

[0040] The humanoid robot model also includes the construction of NVIDIA Omnigraph, task flow design and feedback control. NVIDIA Omnigraph is used for communication between the virtual environment and the external ROS environment. Through Omnigraph, the virtual robot can exchange data with external systems (such as ROS); Task flow design: The robot control algorithm is responsible for the logic of the robot's task execution, including grabbing samples, pouring reagents, and operating heating equipment. Omnigraph serves as a communication bridge to help the robot obtain task instructions and data from the external environment in the virtual environment and pass these instructions to the robot for execution; Feedback control: Through Omnigraph, Isaacsim publishes sensor information in the virtual environment to the external environment, including IMU information, image information and robot status information to the control algorithm. Combined with the feedback information from the sensor, the control algorithm adjusts the robot's movements in real time to ensure the continuity and accuracy of the movements, allowing the robot control algorithm to obtain specific environmental information and robot status information.

[0041] The humanoid robot model also includes ROS2 communication integration, two-way communication and control interfaces. ROS2 communication integration: simulates the interaction between the real robot and the external system, uses ROS2 as the communication middleware between the robot and the simulation environment, and integrates it with IsaacSim to ensure the consistency between the simulation and the real system; Two-way communication: Through ROS2, the robot receives task instructions from the control system and feeds back its internal state and sensor data to the external system, completely simulating the behavior of the real robot communicating with the outside world during the simulation process; Control interface: ROS2 provides a standardized control interface for the robot, so that the motion control, data collection and feedback mechanisms in the simulation are consistent with the real robot. After the simulation is completed, the ROS2 communication part is applied to the actual robot deployment to reduce the debugging time in the actual environment.

[0042] Step 3: Configure the workflow of different robot tasks in the simulation scenario. Set up the typical task flow of the humanoid robot in the simulation scenario, including sampling, analysis, classification and handling tasks in the inspection laboratory. By executing the task sequence in the simulation, obtain the performance data of the algorithm in the experimental task, verify and optimize the robot's control accuracy and task completion efficiency; Step 3: Based on the virtual environment and robot model, design the robot's task flow in the steel inspection process, including the following key operation steps: Sample collection: The robot grabs and takes out the steel sample from the sample container and places it in the designated testing equipment; Reagent addition: The robot accurately pours the required chemical reagents from the reagent bottle and adds them to the sample for chemical reaction; Equipment operation: The robot operates experimental equipment by simulating hand movements, such as starting heaters, controlling mixers, and setting temperature parameters; Result reading: The robot reads the result data displayed by the experimental equipment through the built-in sensor, and transmits the data back to the control center for processing through the simulation software; These operations are simulated in real time using IsaacSim's physics engine to ensure the accuracy of the robot's movements and its adaptability to the laboratory environment.

[0043] Step 4: Collect and analyze data when the robot performs tasks in the virtual environment. Collect and analyze data on the performance of the humanoid robot in different task scenarios during the simulation process, including visual recognition accuracy, time efficiency of path planning, operational stability, collision detection parameters, and comprehensive evaluation of the algorithm performance, and generate an experimental report; The different task scenarios in step 4 include normal working process and abnormal scenario testing. Normal working process: simulate the standard operating process in the steel inspection and testing laboratory to ensure that the robot completes the entire process from sample processing to data reading; Abnormal scenario testing: When unexpected events are involved in the simulation, such as experimental equipment failure, sample dropping, reagent spillage, etc., verify whether the robot can respond appropriately in unexpected situations and avoid further losses.

[0044] The different mission scenarios in step 4 also include extreme working condition testing: simulating complex or extreme environmental conditions, such as high temperature or confined laboratory space, to test the robot's operating performance and stability in these environments.

[0045] Step 5: Adjust the algorithm or control parameters based on the simulation result feedback to improve the robot's adaptability and stability in complex laboratory environments; In step 5, during the simulation operation, the system records each step of the robot operation, including the motion path, operation time, task success rate, and equipment collision. By analyzing these data, the robot's operation bottlenecks and potential risk points are identified. The specific analysis methods include: Path optimization: analyzing the robot's motion trajectory when performing tasks, optimizing its joint motion and path planning, reducing unnecessary movements, and improving work efficiency; Collision detection: record the collision between the robot and the environment or equipment, adjust the action planning and operation accuracy, and avoid equipment damage or robot damage in actual operation; Iterative optimization is performed based on simulation data to adjust the robot's motion control algorithm and task flow to ensure that it can complete laboratory tasks safely and efficiently in a real laboratory environment.

[0046] It is obvious to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, from all points of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention, and any reference signs in the claims should not be regarded as limiting the rights involved.

Claims

1. A humanoid robot simulation method for steel inspection and testing laboratory based on IsaacSim, characterized by: The IsaacSim-based humanoid robot simulation method for steel inspection and testing laboratory comprises the following steps: Step 1: Carry out detailed modeling of the environmental elements of the steel inspection laboratory, carry out detailed modeling of the key components and environmental elements of the steel inspection laboratory, establish a digital model consistent with the actual environment, and deploy the model to the database of the IsaacSim simulation environment to ensure that the simulation scene accurately reflects the operating environment and spatial layout of the real laboratory; Step 2: Integrate the humanoid robot model in the IsaacSim simulation environment and load the control system, including visual recognition, motion planning, and control algorithm modules, so that it can perform specific tasks in the simulation scenario. By setting perception sensors and motion drives for the robot model, the perception sensors include cameras and lidars, so that it has the same environmental perception and motion capabilities as a real robot. Step 3: Configure the workflow of different robot tasks in the simulation scenario. Set up the typical task flow of the humanoid robot in the simulation scenario, including sampling, analysis, classification and handling tasks in the inspection laboratory. By executing the task sequence in the simulation, obtain the performance data of the algorithm in the experimental task, verify and optimize the robot's control accuracy and task completion efficiency; Step 4: Collect and analyze data when the robot performs tasks in the virtual environment. Collect and analyze data on the performance of the humanoid robot in different task scenarios during the simulation process, including visual recognition accuracy, time efficiency of path planning, operational stability, collision detection parameters, and comprehensive evaluation of the algorithm performance, and generate an experimental report; Step 5: Adjust the algorithm or control parameters based on the simulation result feedback to improve the robot's adaptability and stability in complex laboratory environments.

2. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 1, characterized in that: The step 1 creates a virtual environment of a steel inspection laboratory on the IsaacSim simulation platform, the environment including the equipment and tools in the laboratory, and ensures realistic physical interaction between the equipment in the virtual environment and the humanoid robot by configuring corresponding physical properties for the equipment and tools in the laboratory; Equipment layout: According to the equipment layout of the actual steel inspection and testing laboratory, the specific location of the equipment is set in the simulation environment to ensure that the simulation environment is consistent with the real laboratory; Detailed modeling of tools: Detailed modeling and functional configuration are performed for the tools that the robot needs to use in the experiment so that the robot can operate effectively in the simulation.

3. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 1, characterized in that: The humanoid robot model in step 2 includes kinematic and dynamic models, sensor simulation and robot physical characteristics. The kinematic and dynamic models include its skeleton, joints, actuators and sensors to ensure that the model imitates human joint movements and flexibly performs complex laboratory tasks such as grasping, carrying and operating instruments. Each joint is simulated using a high-precision physics engine to ensure that the robot's behavior in the simulation environment is consistent with the actual hardware.

4. Robot physics: Set the robot’s physical properties, including mass, center of gravity, and inertia, so that the robot can exhibit realistic motion in a virtual environment and simulate the loads and dynamic behaviors of actual work.

5. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 3, characterized in that: The sensor simulates an integrated visual sensor and an IMU sensor to capture information in a virtual environment and feed it back to the control system to help the robot perform precise operations in the simulation.

6. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 4, characterized in that: The step 2 includes the construction of NVIDIA Omnigraph, task flow design and feedback control. The NVIDIA Omnigraph is constructed: Omnigraph is used for communication between the virtual environment and the external ROS environment. Through Omnigraph, the virtual robot interacts with the external system for data. Task flow design: The robot control algorithm is responsible for the logic of the robot's task execution, including grabbing samples, pouring reagents, and operating heating equipment. Omnigraph serves as a communication bridge to help the robot obtain task instructions and data from the external environment in the virtual environment; Feedback control: Through Omnigraph, Isaacsim publishes sensor information in the virtual environment to the external environment, including IMU information, image information and robot status information to the control algorithm. Combined with the feedback information from the sensor, the control algorithm adjusts the robot's movements in real time to ensure the continuity and accuracy of the movements.

7. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 5, characterized in that: The humanoid robot model in step 2 also includes ROS2 communication integration, two-way communication and control interface. ROS2 communication integration: simulates the interaction between the real robot and the external system, uses ROS2 as the communication middleware between the robot and the simulation environment, and integrates it with IsaacSim to ensure the consistency between the simulation and the real system; Two-way communication: Through ROS2, the robot receives task instructions from the control system and feeds back its internal state and sensor data to the external system, completely simulating the behavior of the real robot communicating with the outside world during the simulation process; Control interface: ROS2 provides a standardized control interface for the robot, so that the motion control, data collection and feedback mechanisms in the simulation are consistent with the real robot. After the simulation is completed, the ROS2 communication part is applied to the actual robot deployment to reduce the debugging time in the actual environment.

8. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 1, characterized in that: The step 3 designs the task flow of the robot in the steel inspection process based on the virtual environment and the robot model, including the following key operation steps: Sample collection: The robot grabs and takes out the steel sample from the sample container and places it in the designated testing equipment; Reagent addition: The robot accurately pours the required chemical reagents from the reagent bottle and adds them to the sample for chemical reaction; Equipment operation: The robot operates experimental equipment by simulating hand movements; Result reading: The robot reads the result data displayed by the experimental equipment through the built-in sensor, and transmits the data back to the control center for processing through the simulation software; These operations are simulated in real time using IsaacSim's physics engine to ensure the accuracy of the robot's movements and its adaptability to the laboratory environment.

9. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 1, characterized in that: The different task scenarios in step 4 include normal working process and abnormal scenario testing. Normal working process: simulate the standard operating process in the steel inspection and testing laboratory to ensure that the robot completes the whole process from sample processing to data reading; Abnormal scenario testing: When unexpected events are involved in the simulation, verify whether the robot can respond appropriately in unexpected situations and avoid further losses.

10. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 8, characterized in that: The different task scenarios in step 4 also include extreme working condition testing: simulating complex or extreme environmental conditions to test the robot's operating performance and stability in these environments.

11. The method for simulating a humanoid robot in a steel inspection laboratory based on IsaacSim according to claim 1, characterized in that: In step 5, during the simulation operation, the system records each step of the robot operation, including the motion path, operation time, task success rate and equipment collision situation, and identifies the robot's operation bottlenecks and potential risk points by analyzing these data. The specific analysis methods include: Path optimization: analyzing the motion trajectory of the robot when performing tasks, optimizing its joint motion and path planning, reducing unnecessary movements, and improving work efficiency; Collision detection: record the collision between the robot and the environment or equipment, adjust the action planning and operation accuracy, and avoid equipment damage or robot damage in actual operation; Iterative optimization is performed based on simulation data to adjust the robot's motion control algorithm and task flow to ensure that it can complete laboratory tasks safely and efficiently in a real laboratory environment.

Images