Immune evolutionary path planning with instance-learning for mobile robot under changing environment

This paper has present an immune evolutionary planning and negating algorithm with instance-learning for mobile robot under changing environment, which combines immune principle in life science with instance-learning into evolutionary algorithm. Experiences (excellent individuals) in elapsed evolutionary process are stored by instances, and by means of instance-learning immune evolutionary algorithms can quickly plan global-optimal path. Then roles of instance-learning and immune are analyzed from mathematical analyses as well as simulating experiments.