Leader-Follower Adaptive Formation Control without Observation of Agent's Velocity Using Wavelet Neural Network
Masanao Obayashi, Takeshi Aridome, Takashi Kuremoto, Shingo Mabu · 2017
This paper proposes an intelligent formation control method of the leader and follower agents without observing velocity of the agents using wavelet neural networks with auto-structuring mechanism. In the proposed method, each agent can follow the leader, exchanging only information of positions (without velocities) of the communicable agents including the leader, and taking a predefined formation. In real world, the method that doesn't need a lot of information to make cooperative behaviors is very useful for such cases of environments existing weak communications. In computer simulations, it is verified that the proposed method has higher performance than conventional methods and is useful in the points of the leader-following formation control without velocity and of the changeable environment.