REAL-TIME VISUAL TARGET IDENTIFICATION AND TRACKING VIA UNMANNED GROUND VEHICLE (UGV)

Abstract

Vision-based Autonomous Robots field is becoming rapidly popular, according to the artificial intelligence revolution. The proposed system is a composed Unmanned Ground Vehicle vision-based target tracking robot prototype. Target tracking is useful in multiple real-life issues such as the assistance and security fields. Acquitted visual input processing is applied through using OpenCV library. The object detection stage of UGV system, is done based on a pre-built deep learning object detection model. YOLOv3-tiny is used to detect objects, which is a light computation-cost version comparing to original YOLO models, and the other complex deep-learning networks. Object to track is specified to be a human only. The target tracking algorithm is based on a sequence of mathematical equations with Region of Interest and stream’s frame coefficients. Coefficients refer to the values of locations according to x-axis and y-axis of the frame. A simple mathematical technique is used for the delayed feedback issue. The locomotion of UGV is based on transmitted commands from algorithm to the motors through local network connection. Ultra-sound technique is used for collision avoidance. Robotic eye-based face tracking is a subsystem. In the subsystem, the face detection stage in Robotic eye-based face tracking subsystem is done using Harr-like cascade. The locomotion algorithm is based on various trigonometry rules using pan/tilt and servos means. Innovative methods are followed for performance measurement and comparison. The results show, an autonomous behavior, streamlined and accurate locomotion of tracking. Keywords: Autonomous Robot, UGV, Object Tracking, Target Tracking, Human Following, Object Detection, YOLOv3, COCO dataset, Face tracking, Robotic eye, pan/tilt