Research On Human-machine Collaborative Control Strategy Of Manned Legged Mobile Platform

Compared with conventional wheeled and tracked ground mobile platforms,legged mobile platforms have stronger terrain adaptability and better stability which are widely used in disaster rescue,material transportation,geological exploration and other field operations.However,the current autonomous decision-making algorithm developed for the manned mobile platform is difficult to ensure its safety and stability on complex terrain.If the platform is completely controlled by the driver,problems denoted by high handling difficulty and heavy load will decrease the handling efficiency.Hence,to realize the complementary advantages of human-machine cooperation and address the problems above,the design of a human-machine cooperative control system which organically integrates the decision-making ability of human drivers for unknown complex situations with the powerful computing ability of intelligent systems is undoubtedly an effective solution.In this dissertation,the leg kinematics model is established according to the characteristics of the platform configuration.The foot trajectory is planned and the stability of the platform is analyzed;According to the driver manipulation habits and platform characteristics,a set of fully functional legged mobile platform manipulation system,and a simple human-machine cooperative driving mode with high computational efficiency are designed to lay a solid foundation for the subsequent research of the human-machine cooperative control strategies.Then,the terrain is classified,and a fuzzy control system designed for quantifying the terrain complexity is proposed.A criterion for quantitatively evaluating the reliability of drivers is proposed.Fully considering driver reliability and terrain complexity,a dynamic allocation strategy of driving rights based on two-layer fuzzy control system is proposed to achieve dynamic allocation of driving rights.The experimental results show that the strategy can reasonably and dynamically allocate driving rights according to the driver reliability and terrain complexity.Furthermore,an autonomous command optimization strategy based on the fusion of the terrain information and the optimal solution of cost function is proposed for the platform,and a human-machine intention analysis method is designed.Based on the human-machine intention analysis and automatic driving factor,a weighted human-machine command fusion strategy is designed.The effectiveness of the command fusion strategy is verified through simulation experiments.Finally,a driving simulation experiment system with high driving reality is established to realize driving environment visualization and driver manipulation data collection.Human-in-loop driving experiments are carried out,and the experiment results are discussed with complete data analysis.The experiment results show that the control strategy designed in this paper can achieve variable weight human-machine cooperative control.