Compliance Control and Simulation of a Next-Generation Chewing Robot

摘要

The aim of the development of the chewing robots is to study the biomechanics of chewing process or produce the bolus for further research by in vitro experiments. Some chewing robots are open loop controlled and couldn't behave like a human. However, to make the chewing robots more bin-mimic, force feedback based closed loop control is required. To improve the efficiency of the in-vitro chewing experiments, a next generation chewing robot with cam-follower based multi-chambers was developed previously, which can be used test the effect of different chewing parameters on the chewing performance and food texture changes simultanously. In this paper, modified admittance control based on the force feedback was applied to the chewing robot. At first, to improve food particle size reduction to a distribution which closely matches that produced by human and protect the components of the robot from breakage or impact damage, a spring mass based system, called remote center compliance (RCC), was developed and applied to the robot to control the chewing force. Then, based on the knowledge of human chewing process, power-based algorithm, and modified admittance control method were used as the feedback to adjust the angular velocity of the motor and corresponding chewing velocity of the teeth. The performance of the control model was also simulated, results showed that the model is reliable to control the robot and make it bio-mimic.