Various control methods based on visual information have been studied for mobile robots for years. Among them, image-based control is said to have high robustness against camera calibration errors. A drawback of image-based control is that the transitional behavior of the system is difficult to predict because it is not designed in the Cartesian space but in the image plane. Especially, unfavorable transitional behavior may cause the loss of target from the sight. In this paper, a design method of the image-based stabilizing control for a nonholonomic mobile robot is proposed not to go beyond the targets by using the state transformation. The effectiveness of the proposed method is confirmed by experiments with a two-wheeled mobile robot.
This paper considers a continuous-discrete (C-D) nonlinear filtering problem by using the divided difference method, and develops C-D versions of the divided difference first-order (DD1) filter and the divided difference second-order (DD2) filter by extending the results due to Schei , Norgaard, et al  and Simandl, et al . Since Jacobians are not needed in designing C-D DD1 and DD2 filters, these filters can be applied to the filtering problem for any C-D nonlinear systems, to which the C-D extended Kalman filter (EKF) is difficult to apply. Simulation results show that the performance of the proposed C-D DD2 algorithm is comparable to that of the C-D unscented Kalman filter (UKF), though the performance of the C-D DD1 filter is slightly inferior to them.
This paper presents a distributed control method of multi-robot systems for performing the mass games, i.e., to display a given image as a formation of a number of mobile robots. The proposed method is based on the combination of a coverage control method and halftone image processing. The performance is demonstrated by numerical simulations with several standard images.