| The mobile robot path planning is a key research area in robotics.This paper proposes a hybrid path planning method in view of the partly known environment and unknown static obstacles.At first,the particle swarm optimization(PSO)algorithm is directly used in the environmental model by improving the encoding to search out an appropriate path.And then a local obstacle avoidance algorithm based on fuzzy neural network associated with the global path is presented,which approach the nearly optimal path.Firstly,when the global path planning is beginning,this paper uses the MAKLINK graph to build the environmental model firstly,and then through improving the coding math to make particle swarm algorithm directly used in the model to calculate an appropriate path,and continuing use the PSO algorithm to optimize this path.In addition,the random factor in the algorithm can avoid it traps into local optimum.The simulation result shows that the method can decrease the complexity of algorithm and calculate the appropriate path in shortest time compared to the traditional method.As the local path planning of the hybrid approach,a fuzzy neural network structure is used in the local path planning based on the demand,and training the network structure sequentially.Based on these theories,this paper disposes the environmental information and simulates the sensor function so as to the simulation result is near to the real environment.The simulation results in different environment demonstrate that this method is effective and has the strong robustness.Finally,the hybrid method is proposed.The method is that regarding the path-point of the global optimal path as the target point of the local path planning,which will make the robot searched out a local optimal point quickly,and moved to the terminal point associated with the global path.This method will reduce more time and save more cost of the planning.The simulation result shows that this method is feasible. |
PDF Full Text Request