The First IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics, 2006. BioRob 2006.
DOI: 10.1109/biorob.2006.1639126
Full text: Unavailable
A state observer for estimating and predicting the internal state of polypyrrole conducting polymer actuators is presented. Polypyrrole actuators can produce large stresses (>5 MPa) with large associated strains (>12%), which leads to a high energy density. However, care must be taken since conducting polymers degrade quickly when exposed to overcharged or over-strain conditions. Unfortunately, charge and other internal state variables relevant to degradation of the actuator material are not directly measurable in real-time control systems. To cope with this difficulty, this paper presents the design of a dynamic state observer to estimate the internal state of the polymer actuator from available sensor information. First, a reduced-order model of a polypyrrole conducting polymer actuator is obtained. A state observer is then designed based on the dynamic model. Subsequently, a method is presented for real-time identification of key model parameters from the measured data. Lastly, a novel control system is presented which utilizes the estimated internal states of the actuator to slow its performance degradation. Preliminary experiments and simulation are presented to demonstrate the proposed approach