Kenichiro Nonaka

Slip measurement and vehicle control for leg/wheel mobile robots using caster type odometers

In this paper, we propose a velocity control method for leg/wheel type robots by using tire side slip angle and tire slip ratio. Side slip angle and slip ratio is measured by caster type odometers which are comprised of directional sensors and travel distance sensors. The odometer system has redundant four encoders to measure three dimensional velocities: planar two dimensional and rotational movements. We reduce the error between target velocity and actual velocity by feedback control based on both tire slip angle and slip ratio. This method can improve stability and accuracy of movement of vehicles. The experimental vehicle is an omnidirectional vehicle which has legs comprised of three joints. Experimental results of circular trajectory running show effectiveness of the proposed control method.

Optimal online generation of obstacle avoidance trajectory running on a low speed embedded CPU for vehicles

Considering obstacle avoidance for mobile robots, it is effective to generate optimal trajectory dynamically in terms of safety and efficiency. Since it usually requires iterative calculation to get an optimal trajectory, however, high computational cost becomes a problem for practical application, especially for low performance computers like industrial embedded CPUs. In this paper, we propose a low cost trajectory optimization method which is applicable to embedded CPUs. This method suppresses the computational cost through introduction of a low order polynomial function and proper boundary conditions to generate continuous trajectory. In order to see the feasibility, we applied proposed method to a wheeled robot which is equipped with an embedded CPU and a laser range finder (LRF) to show the performance of obstacle avoidance experiments.

Experiments of real-time numerical sliding mode control for vehicles

Since the longitudinal, lateral, and yaw dynamics of front steering vehicles are represented by nonlinear coupled equations, it is not easy to derive the explicit nonlinear controller of both steering angle and driving force for the perfect path-following control. In this paper, a novel and robust sliding mode controller for vehicles with coupled full dynamic model is presented. Comparing with the conventional vehicle controllers, this controller deals with coupled longitudinal, lateral, and yaw dynamics simultaneously, while ensuring stability and robustness for the uncertainty and external disturbances. Since these coupled dynamics are nonlinear and difficult to solve analytically, we use numerical solution which is feasible at real-time computation. In addition, to achieve robustness, this controller includes the sliding mode control to deal with uncertainty of road surface. The advantage of this controller is verified through experiments.

Analysis of time delays for time-state control form systems applied to vehicle motion control using visual feedback

In recent years, automatic running of vehicle type robots are studied actively. There are various kinds of methods: visual feedback is one of them. Issue of this system is that it includes time delays during signal transmission. Since time delays degrade motion performance of vehicle, it is necessary to consider it to recover the performance. Time state control form which we use in this paper is effective to path tracking, however, time delays for that form have not been considered yet. It is necessary to consider delays for stable motion control. In this paper, we analyze stability of vehicle motion in time state control form with time delays.

Development of a Visual Feedback Navigation System using Multiple Image Processing Hosts

Abstract In recent years, automatic running of the vehicle robots is studied actively. There are various kinds of methods; visual feedback control is one of them. In our laboratory, ten cameras set below the ceiling capture vehicles states; then each image are processed by host computer and vehicles are controlled. At the moment, the experimental course is divided into two areas. Since each area is covered by independent host computers which process four or six cameras respectively, vehicles cannot run in these two areas; the trajectory that vehicle run is limited. The aim of this study is to introduce the development of communication system between hosts. The communication between host computers is realized using LAN. We also designed prediction of image processing area to prevent losing sight of vehicle. We verified advantage of this system through experiments. Moreover, since this system includes time delay, we identified delay factors. In addition, we verified delay influences through experiments and showed advantages of this system for education and research.