PROJECT TITLE :

A Robust Visual Human Detection Approach With UKF-Based Motion Tracking for a Mobile Robot

ABSTRACT:

Strong tracking of a person's in a very video sequence is an essential requirement to an increasing number of applications, where a robot wants to interact with an individual's user or operates in a human-inhabited setting. This paper presents a sturdy approach that enables a mobile robot to detect and track somebody's using an onboard RGB-D sensor. Such robots could be used for security, Surveillance, and assistive robotics applications. The proposed approach has real-time computation power through a unique combination of new ideas and well-established techniques. Within the proposed technique, background subtraction is combined with depth segmentation detector and template matching method to initialize the human tracking automatically. A unique concept of head and hand creation based mostly on depth of interest is introduced during this paper to track the human silhouette in an exceedingly dynamic environment, when the robot is moving. To form the algorithm robust, a series of detectors (e.g., height, size, and form) is used to tell apart target human from different objects. As a result of of the comparatively high computation time of the silhouette-matching-based mostly methodology, a confidence level is outlined, which permits using the matching-based method only where it is imperative. An unscented Kalman filter is employed to predict the human location in the image frame to keep up the continuity of the robot motion. The efficacy of the approach is demonstrated through a true experiment on a mobile robot navigating in an indoor environment.