Research On Local Planning Of Amphibious Robots Under Emergencie

As an inanimate carrier,robot is playing an increasingly important role in dangerous and complex environment.Compared with traditional robot that can only work in a single environment,such as ground robot,aerial robot,and etc.,triphibious robot has the features of wider range of movement and stronger adaptability to complex environment.When a triphibious robot encounters obstacles,pits,blockage and other emergencies,it needs to plan a safe and efficient trajectory independently in the three environmental domains of water,land and air.When it is difficult to move in a certain environment domain and the robot is required to move across domains,it needs to decide the optimal working mode in the environment with unknown local information.Therefore,taking the triphibious robot under emergency as the research object,researches on cross-domain path planning algorithm and decision-making algorithm of working mode switching have been carried out.Firstly,to meet the needs of emergency response,the Bayesian probability updating method is adopted to construct a two-dimensional local grid map with the information of obstacle areas,blocked areas and pit areas in the ground environment.Also,a local emergency planning algorithm that integrates the dynamic window method with the safe area search strategy is proposed,which enables the robot to avoid emergencies in the ground working mode.Secondly,to solve the problem that traditional ant colony algorithm cannot plan crossdomain trajectories and the path evaluation index is single,in the three-dimensional grid environment model,fuzzy rules are used to determine the scope of the local replanning window when the triphibious robot encounters sudden obstacles.The initialized pheromones of ant colony are redistributed according to the different characteristics of water,land and air environment.The search heuristic function is improved according to the factors of emergency safety,working condition switching safety,time-consuming of nodes and energy consumption of working mode.The results of simulation experiment show that the improved triphibious ant colony algorithm based on fuzzy window proposed in this paper can generate collision-free cross-domain trajectories,and has the advantages of fast convergence speed,low energy consumption and short flight time.Then,to make robot choose the optimal working mode in local unknown environment,a decision-making method based on multi-attribute cross-domain switching is proposed.The working mode switching scheme of the triphibious robot is constructed.The subjective weight is modified by combining subjective weight with objective weight.State machines which can autonomously switch modes in water,land,air and emergency environment are designed to realize autonomous switching of robot motion states.The simulation results verify the effectiveness of the multi-attribute cross-domain handover decision-making method.Finally,the experimental platform of triphibious robot is built.Also,the path-planning experiment and decision switching experiment are separately carried out on the basis of robot autonomous control.Experimental results verify the effectiveness of the proposed method.The results of path-planning experiment show that the planning algorithm adopted by the triphibious robot can build cross-domain paths and avoid local obstacles.The results of decision switching experiment show that the triphibious robot can autonomously decide the optimal working mode in unknown environment and complete the switching.