Visual tracking of a moving target by a camera mounted on a robot: a combination of control and vision

Active Tracking System with Rapid Eye Movement Involving Simultaneous Top-down and Bottom-up Attention Control

“…This event driven control turned out to be helpful for long-duration-tracking. Appropriate selection of control strategy is important to maximize performance (Papanikolopoulos et al, 1993), in terms of speed, stability, and robustness of active tracking systems. In this paper, we explored co-realization of saccadic and smooth pursuit using hybrid multi-rate control system (i.e., slow FF-velocity control and fast FB-position control).…”

Real-Time Tracking of Moving Objects with an Active Camera

“…The error in the image space is transformed to the joint space where the regulation is performed by six PD controllers, one for each joint angle. Papanikolopoulos et al [26] use the optical flow in the center of the image to track an object. Four different control methods (LQG, pole assignment with DARMA and ARMAX models, and PI) are compared, with special emphasis on the disturbance treatment.…”

Optimal Importance Sampling for Tracking in Image Sequences: Application to Point Tracking

“…In particular, one has to face difficult and ambiguous situations generated by cluttered backgrounds, occlusions, large geometric deformations, illumination changes or noisy data. To design trackers robust to outliers and occlusions, a classical way consists in resorting to stochastic filtering techniques such as Kalman filter [13,15] or sequential Monte Carlo approximation methods (called particle filters) [7,10,11,16].…”