Study Of Multi-Objective Task Planning Of Space Manipulator Under Multi-Constraint

With the rapid development of the international space industry and space exploration,the task of space operation has become more and more complicated.Using space manipulator to replace or auxiliary astronaut to complex varieties of on-orbit tasks can effectively increase the economic benefit and safety of spacecraft in orbit service,therefore,space manipulator is a significant role in future space exploration.Space work environment information is difficult to obtain,many factors are uncertain,and the spacecraft and command center of teleoperation time delay problem increasingly prominent,that put forward new requirements of autonomous decision-making ability for space manipulator.Limited by the complex space environment and operation,task like space maintenance that cannot be achieved by a single action has a strong binding,it need to meet such as joint angle and angular velocity limits under the premise first and then remove the old parts to replace the new.In addition,due to the limitation of space resources,reasonable configuration of spacecraft resources is also important.This paper studies the multi-constraint multi-objective spatial manipulator task planning method.The research content is as follows:Firstly based on the task characteristics of space manipulator,space manipulator general framework is put forward and mathematical model of mission planning method is established on the basis of which Combing space manipulator preform,task constraints,and optimization goals.The framework and task mathematical model can describe multi-constraint multi-objective manipulator mission planning processes,and description mathematically for the mission planning process.Then researching space manipulator multi-constraint multi-objective task execution strategy,based on the planning and improved simulated annealing algorithm,and combining the task characteristics of space manipulator,putting forward a method of planning steps controllable figure development to achieve mission planning results from single to the collection.Further,this paper proposes a multi-constraint multi-objective task planning solution extraction method,based on task sequence set,in the constraint space by integrating different target planning mission planning execution strategy.The method fully combines the task characteristics of the space manipulator with intelligent graph planning features,implements the more constrained multi-objective autonomous mission planning process.Then aiming at the mobile task of space manipulator,a space manipulator path decomposition method based on heuristic algorithm is proposed.On this basis,combining with the path of learning strategy and path planning based on neural network time cost estimation method,this paper proposes a mobile mission planning method based on the planning knowledge base.The method that introduces learning mechanism in the heuristic path planning algorithm can be achieved under the premise of considering multi-constraint multi-objective and can improve the speed of task planning with the ability of self-learning.Finally experimental research is carried out on multi-constraint multi-objective space manipulator task planning method.In order to verify the application value of theory in this article in the research field of the manipulator,in view of the typical manipulator operation task,carrying out multi-constrained multi-objective manipulator task planning method simulation and actual product experiment.By analyzing the corresponding experimental results,the correctness and validity of the algorithm are verified.