Immune Network Algorithm Based on Improved APF for On-Line Dynamic Planning

To solve the problem of on-line dynamic planning of mobile robots in unknown environments, inspired by the mechanism of idiotypic network hypothesis, a hybrid immune network algorithm (HINA) is proposed. To improve the planning efficiency of immune network algorithm (INA) and realize optimal on-line dynamic obstacle avoidance, a new adaptive artificial potential field (AAPF) method is presented by using modified potential field. The vaccine is extracted according to the planning results based on AAPF method, and the instruction definition of robot is initialized through vaccine inoculation, which improve the planning efficiency of INA. When the robot meets with moving obstacles during the path planning, the AAPF method is used for the optimal dynamic obstacle avoidance. Simulation results are presented to verify the effectiveness of the proposed algorithm in unknown environments.