Yanan Huang; Peng Cheng; Chia-Hsiang Menq
This paper presents the design of an adaptive observer that is implemented to enable real-time dynamic force sensing and parameter estimation in an optically trapped probing system. According to the principle of separation of estimation and control, the design of this observer is independent of that of the feedback controller when operating within the linear range of the optical trap. Dynamic force sensing, probe steering/clamping, and Brownian motion control can, therefore, be developed separately and activated simultaneously. The adaptive observer utilizes the measured motion of the trapped probe and input control effort to recursively estimate the probe-sample interaction force in real time, along with the estimation of the probing system's trapping bandwidth. This capability is very important to achieving accurate dynamic force sensing in a time-varying process, wherein the trapping dynamics is nonstationary due to local variations of the surrounding medium. The adaptive estimator utilizes the Kalman filter algorithm to compute the time-varying gain in real time and minimize the estimation error for force probing. A series of experiments are conducted to validate the design of and assess the performance of the adaptive observer.
DOI
This paper presents the design of an adaptive observer that is implemented to enable real-time dynamic force sensing and parameter estimation in an optically trapped probing system. According to the principle of separation of estimation and control, the design of this observer is independent of that of the feedback controller when operating within the linear range of the optical trap. Dynamic force sensing, probe steering/clamping, and Brownian motion control can, therefore, be developed separately and activated simultaneously. The adaptive observer utilizes the measured motion of the trapped probe and input control effort to recursively estimate the probe-sample interaction force in real time, along with the estimation of the probing system's trapping bandwidth. This capability is very important to achieving accurate dynamic force sensing in a time-varying process, wherein the trapping dynamics is nonstationary due to local variations of the surrounding medium. The adaptive estimator utilizes the Kalman filter algorithm to compute the time-varying gain in real time and minimize the estimation error for force probing. A series of experiments are conducted to validate the design of and assess the performance of the adaptive observer.
DOI
No comments:
Post a Comment