Path Planning Based on the Extension Human-Simulated Strategy in Unknown Environments

This paper applies the extension strategy to path planning of a mobile robot and proposes a new algorithm. This new method introduces temporary aims in obstacle avoidance and imitates human´s path selection in an unknown environment. So information about the environment is efficiently compressed and the modeling of complicated environment can be avoided in real time calculation. The evaluation function based on safe distance and correlative functions makes the selected path smoother, and lowers the standards of the robot´s self-control and the sensor´s measurable precision. Due to robustness of human imitating strategy, oscillation and local-minimum in other traditional methods can be efficiently avoided. Both experiments and simulations show that this method has good real-time performance and work better in path planning than other methods.