Font Size: a A A

Key Technologies Research On Hand Physiological Tremor Testing

Posted on:2014-08-07Degree:MasterType:Thesis
Country:ChinaCandidate:H Y DaiFull Text:PDF
GTID:2268330425965915Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Physiological tremor is an involuntary movement characterized by regular oscillationsof one or several body parts. It is difficult for operation in micromanipulation because theamplitude of physiological tremor is equal to the operation accuracy. The active tremorcompensation system has a simple principle and is inexpensive, while in which the tremorshould be accurately measured. A tremor test system is designed in this paper, based on theprinciple of inertial measurement, in order to accomplish arm tremor measurement.The tremor test system contains two parts, signal measurement and processing. Thetremor signal is measured via non-gyro inertial measurement unit, in order to solve theproblem of the bottlenecks in optical measurement practical application and low accuracy ofMEMS gyroscope. Inertial Measurement Unit is designed in9accelerometers configurationmode to avoid solving differential equations. To improve the measurement accuracy, theintegral-evolution fusion algorithm is adopted in solving the angular velocity, avoiding ac-cumulated error in solving the angular velocity by integration method. Strap-down matrix isadopted for coordinate transformation of physical quantities, and the matrix is solved via thequaternion, making it easy to achieve on a real-time system.The signal processing containserror analysis, the tremor signal extraction and modeling. Error sources of the measurementsystem, especially the frequency characteristics of the systematic and random errors, areanalyzed. Parameter identification for random error is conducted by Allan variance method,followed by modeling via first-order Gauss-Markov model. On basis of the frequency cha-racteristic analysis for the noise and tremor signal, the noise is eliminated via filtering me-thod, and signal-noise separation is accomplished. According to the similarity of the tremorand the sinusoidal signal, fitting by Fourier and its derivative algorithm and forward predic-tion by auto-regressive model for the tremor signal are conducted.Parameter identification of the output signal is completed by the Allan variance analy-sis after tests for accelerometers under static conditions. Then, tremor signal extracting me-thod is verified by tests for accelerometers under dynamic conditions. Finally, tracking per-formance of the target locus is verified by dynamic tests for the inertial measurement unit.Tremor test system, which is based on the principle of inertial measurement, has asmaller size and costs less. It can be used as the signal input terminal for the active tremor compensation system. Study on he tremor signal can provide reference for the developmentof other types of shock suppression system and research on the mechanism of the tremor.
Keywords/Search Tags:Physiological tremor, inertial measurement unit, Fourier fitting, Allan variance, Gauss-Markov random process
PDF Full Text Request
Related items