Research On The Virtual Task Training System For Space Manipulator

In the process of the construction and operation of the Chinese Space Station,the efficiency and safety of on-orbit operations are being improved,and the Space robotic manipulator needs to be used to assist or replace astronauts in completing on-orbit operations,which has led to the Space robotic manipulator becoming a key piece of equipment essential to the construction and operation of the Chinese Space Station.Due to the special space environment,complex and heavy in-orbit operation tasks,the space robotic manipulator ground controllers are flight control personnel,putting forward higher requirements,and the space robotic manipulator structure is complex,the lack of ground physical training platform,the traditional job training system and mechanism is difficult to adapt to the space robotic manipulator such a complex system training needs,so there is an urgent need to build in line with the current flight control job training task needs The virtual training system for space robotic manipulator tasks.The efficiency of job training is improved and the cost of job training is reduced,which has important engineering value and practical significance for the construction of the Chinese space station and the development of the robotic manipulator mission virtual training field.This paper intends to carry out research work on the virtual job training mechanism,robotic manipulator collision detection and contact force feedback in virtual scenarios for the robotic manipulator task virtual training system,as follows.First,the virtual reality-based virtual training system software architecture design for space robotic manipulator tasks needs to be carried out.The functional requirements and performance requirements are analysed through the virtual training system.The required functional modules of the training system and the corresponding performance indicators are determined;based on the classical MVC system architecture pattern,the view layer,control layer and model layer of the training system are designed in conjunction with the results of the requirements analysis of the training system,reducing the coupling between the system modules and improving the flexibility and scalability of the system based on the idea of encapsulation layering;the functional modules of the system are divided according to the requirements analysis and architecture design of the virtual training system The system's functional modules are divided according to the requirements analysis and architecture design of the virtual training system,and its workflow is designed;the typical in-orbit task characteristics of the space robotic manipulator are analysed,and the assessment index and assessment process of each task are formulated.Secondly,the research on the collision detection method of robotic manipulator in the virtual scene is carried out.Based on the hybrid enclosing box octree,the collision detection algorithm is studied,and the space robotic manipulator model and the space station module model are pre-processed,which leads to the construction of the hybrid enclosing box octree of the model point cloud file.The collision detection algorithm based on the depth-first search algorithm and the two-way breadth search algorithm are designed for the single configuration of the robot manipulator and the continuous motion of the robot manipulator,respectively,to improve the computational efficiency of the collision detection algorithm in the virtual training system.Once again,research on a robotic manipulator force feedback system based on the spring-damping model needs to be completed,which allows for the overall architecture design,force feedback device selection and workflow design of the force feedback system for the synchronisation problem of graphics rendering and force rendering in the virtual training system.Collision detection algorithms are combined with a spring-damping model,which calculates the contact forces when the models come into contact;experiments are designed and carried out to verify the effectiveness and robustness of the force feedback system and force feedback algorithms.Finally,for the validity of the virtual training system for job training to be verified,the virtual training system for space robotic manipulator tasks developed in this paper,which can be applied in the field of robotic manipulator operation job training needs to be demonstrated by building a virtual training system with the sky and robotic manipulator system as the object,and for the virtual scenario designed in this paper for the space robotic manipulator collision detection algorithm and force feedback system,where the corresponding operation task training experiments are designed,which can verify the accuracy and effectiveness of the algorithm and training system.