Yasuaki Kaneda

Development of an artificial muscle linear actuator using ionic polymer–metal composites

We are developing an artificial muscle linear actuator using ionic polymer-metal composites (IPMC)—electro-active polymers that bend in response to electric stimuli—and the goal of our study is applying the actuator to robotic applications, especially to a biped walking robot. In this paper, we will describe the structure of the actuator and an empirical model of the actuator which has two inputs and one output, and whose parameters are identified from input-output data. Based on the empirical model, basic experiments and position control of the linear actuator are demonstrated. Then, we consider walking control of a small-sized biped walking robot. In the application we assume that the developed actuators are connected both in series and in parallel to a joint of the walking robot so that the actuators supply enough torque to the robot, and that they are stretched and compressed enough. It is shown throughout simulations that the biped walking robot with the actuators can walk on level ground with a period synchronized with the period of the input signal.

Development of artificial muscle actuator using ionic polymer with its application to biped walking robots

We are developing an artificial muscle linear actuator using ionic polymer-metal composites (IPMC) which is an electro-active polymer that bends in response to electric stimuli and the goal of our study is to apply the actuator to robotic applications especially to a biped walking robot. In this paper, we will describe the structure of the actuator and an empirical model of the actuator which has two inputs and one output, and whose parameters are identified from input-output data. Based on the empirical model, we demonstrate walking simulations of a small-sized biped walking robot. In the numerical simulation we assume that the developed actuators are connected both in series and in parallel to a joint of the walking robot so that the actuators supply enough torque to the robot and that they are stretched and compressed enough. It is shown throughout the simulation that the biped walking robot with the actuators can walk on a level ground with a period synchronized with a period of input signal.

Design method of robust Kalman filter via ℓ 1 regression and its application for vehicle control with outliers

In many cases, outliers are contained in sensor signals, and these deteriorate performances of control systems, e.g., UAV and UGV using non-contact sensors. Many reduction methods of the outliers have been proposed. One of the methods is robust Kalman filter (RKF) via ℓ1 regression. The method is easy to implement and compute due to a simple structure and convex optimization problem, so the method attracts many attentions. However, parameters of the method are designed by heuristic methods. In this paper, we propose a design method of RKF via ℓ1 regression. We show that statistics of Gaussian noise determine the parameters of RKF, and we can design the parameters systematically. Then, we apply the method to a velocity estimation and control of a two-wheeled vehicle with outliers. Effectiveness is demonstrated by some numerical simulations.

Robust unscented Kalman filter via l 1 regression and design method of its parameters

In this paper, we propose a robust unscented Kalman filter (RUKF) using l1 regression and a new design method of its regularization parameters. Generally, the regularization parameters in l1 regression are designed by heuristic methods, so the parameters have no physical senses. However, in our design method, it is shown that statistics of Gaussian measurement noise determine the parameters of the RUKF, and we can design the parameters systematically. The proposed RUKF is applied to a state estimation of a two-link manipulator with outliers, and the effectiveness is demonstrated by numerical simulations.

Fast Algorithm of Robust Kalman Filter via l1 Regression by a Closed Form Solution

Abstract Robust Kalman filter (RKF) via l 1 regression is a linear filter for non-Gaussian measurement noise, and it can be formulated as a l 1 optimization problem. Generally, the optimization problem cannot be solved analytically, and some numerical iterative methods are needed. This paper proposes a closed form solution of RKF via l 1 regression by an approximation of its optimal solution and it gives a fast algorithm. The approximated solution can be calculated by upper and lower bounds of the optimal solution. Moreover, a bound of an estimation error of the approximated solution can be analyzed. Some numerical simulations demonstrate the effectiveness of the proposed algorithm.

Design method of robust Kalman filter for multi output systems based on statistics

In many cases, outliers are contained in sensor signals, and these deteriorate performances of control systems, e.g., UAV and UGV using non-contact sensors. One of reduction methods of the outliers is robust Kalman filter (RKF) via l1 regression. The method is easy to implement and compute due to a simple structure and convex optimization problem, so the method attracts many attentions. In this paper, we propose a new design method of RKF via l1 regression for multi output systems. It is shown that statistics of Gaussian noise determine the parameters of RKF, and we can design the parameters systematically. RKF with the proposed design method is applied to a two-wheeled vehicle control with outliers, and the effectiveness is demonstrated by numerical simulations.

IPMC Linear Actuator with Re-Doping Capability and its Application to Biped Walking Robot

Abstract We are developing an artificial muscle linear actuator using ionic polymer-metal composite (IPMC) which is an electro-active polymer (EAP) that bends in response to electric stimuli and the goal of our study is applying the actuator to robotic applications especially to a biped walking robot. In this paper, we will describe the structure of the actuator and an empirical model of the actuator, and whose parameters are identified from input-output data. Based on the empirical model, basic experiments and position control of the linear actuator are demonstrated. We consider walking control of a small-sized biped walking robot and investigate changes of the characteristics of the actuator for walking by ionic re-doping chemically. In the application we assume that the developed actuators are connected both in series and in parallel to a joint of the walking robot so that the actuators supply enough torque to the robot and that they are stretched and compressed enough. It is shown throughout simulations that the biped walking robot with the actuators can walk on a level ground with a period synchronized with a period of input signal, and that walking patterns are changed by ionic re-doping.

Robust self-tuning controller under outliers

In this paper, we propose a robust self-tuning controller (STC) under outliers. A parameter update law of a conventional STC consists of a recursive least squares estimation, and the estimation is given by a solution of a minimization problem of estimated errors. In the proposed method, we estimate parameters and outliers explicitly by addition of a l1 regression term to the minimization problem like a robust Kalman filter via l1 regression, and the estimated outliers are removed from measurement outputs in the controller. We also analyze control performances of the proposed method under outliers, and it is shown theoretically that performances in the proposed method with outliers are nearly equal to ones in the conventional STC without outliers. A numerical simulation, in which a controlled plant is a non-minimum phase system, demonstrates effectiveness of the proposed method.

Development and control of hybrid hardware engine simulator

In this paper a novel real-time engine simulator which is combined with partial mechanical hardware system and numerical simulator. In the hardware, the actual piston-clank mechanism is used, and the thrust force is generated by a liner motor. In the numerical simulator, admission and combustion dynamics are calculated, and the motion of the piston is realized by a kind of model matching or following controller. In this paper, we introduce the concept of the proposed simulator system, and explain the control systems.

Reduction of discretization errors of dynamics with variable structure and its realization using FPGA

In this paper, in order to reduce discretization errors of dynamics with variable structures (VS), we propose an improved digital integrator. Use of Richardson extrapolation (RE) and fractional delay (FD) can improve Euler integrator, so we can obtain an improved integrator. However, Euler integrator using RE and FD directly has an infinite gain at a Nyquist frequency, and it is unsuitable for integrations of the dynamics with VS. The proposed method, which is suitable for integrations of the dynamics with VS, consists of a high forward gain and feedback structure with a high-precision differentiator. To realize the high-precision differentiator, RE and a high sampling rate instead of FD are used, and those are implemented by FPGA. Its effectiveness is verified by software simulations, hardware in the loop simulations, and experiments.