SUMMARYThis paper presents a novel line of sight control system for a robot vision tracking system, which uses a position feedforward controller to preposition a camera, and a vision feedback controller to compensate for the positioning error. Continuous target tracking is an important function for service robots, surveillance robots, and cooperating robot systems. However, it is difficult to track a specific target using only vision information, while a robot is in motion. This is especially true when a robot is moving fast or rotating fast. The proposed system controls the camera line of sight, using a feedforward controller based on estimated robot position and motion information. Specifically, the camera is rotated in the direction opposite to the motion of the robot. To implement the system, a disturbance compensator is developed to determine the current position of the robot, even when the robot wheels slip. The disturbance compensator is comprised of two extended Kalman filters (EKFs) and a slip detector. The inputs of the disturbance compensator are data from an accelerometer, a gyroscope, and two wheel-encoders. The vision feedback information, which is the targeting error, is used as the measurement update for the two EKFs. Using output of the disturbance compensator, an actuation module pans the camera to locate a target at the center of an image plane. This line of sight control methodology improves the recognition performance of the vision tracking system, by keeping a target image at the center of an image frame. The proposed system is implemented on a two-wheeled robot. Experiments are performed for various robot motion scenarios in dynamic situations to evaluate the tracking and recognition performance. Experimental results showed the proposed system achieves high tracking and recognition performances with a small targeting error.

No abstract

No abstract

No abstract

This paper presents a vision tracking system to achieve high recognition performance under dynamic circumstances, using a fuzzy logic controller. The main concept of the proposed system is based on the vestibulo-ocular reflex (VOR) and the opto-kinetic reflex (OKR) of the human eye. To realize the VOR concept, MEMS inertial sensors and encoders are used for robot motion detection. This concept turns the camera towards a selected target, counteracting the robot motion. Based on the OKR concept, the targeting errors are periodically compensated, using vision information. The fuzzy logic controller uses sensor data fusion to detect slip or collision occurrences. To calculate a heading angle of the camera accurately, the output of the fuzzy logic controller and the vision information from the camera are combined, using an extended Kalman filter. The proposed vision tracking system is implemented in a mobile robot and evaluated experimentally. The experimental results are obtained as the tracking and the recognition success rate using a mobile robot. The developed system achieved the excellent tracking and recognition performance during slip or collision occurrences under dynamic circumstances.