Distributed Vision-Based Target Tracking Control Using Mobile Robots

Anthony Le and Ryan Clue

Advisors: Dr. Jing Wang, Dr. In Soo Ahn

About the Project

This 2017 Bradley University Electrical Engineering senior capstone project emphasizes a distributed vision-based control system designed for multiple mobile robots to explore the problem of target tracking while maintaining a specified formation among tracking robots. The paper describes related research and the problem formulation for the Kinematic Model of mobile robots utilizing vision-based control for this project. The design of the project’s distributed vision-based control system can be divided into two specific modules: 1) Identifying target and spatial information using images and 2) coordinating motion of multiple robots using autonomous encirclement and formation controls. The project design approach described includes: system/subsystem function definition, QBot2 Mobile Robot hardware and software description, target identification image processing and QBot2 control (vision-based control, encirclement, leader-follower) description. The system integration and testing tasks, experiments and results are also described in the paper as well as project management (resources, task plan and timeline) and the overall design experience from our engineering research project.

The goal of our project is to design distributed vision-based control algorithms for mobile robots and to implement and validate the proposed algorithms. The primary project tasks (research, concept definition, design, and demonstration) tasks included:

1. Design a target identification (detect and locate) module based on RGB image features obtained from a vision sensor

2. Design a target tracking algorithm based on robot model linearization

3. Design a leader-follower formation control algorithm based on depth and image features provided from the target identification module

4. Design a state machine to coordinate target identification module and target control module and communications between Robots

5. Integrate and validate the proposed distributed controls through experimentation in a controlled lab environment

About Us

Anthony Le

Anthony Le is from Peoria, IL. He is currently attending Bradley University with the goal of graduating with a Bachelor of Science in Electrical & Computer Engineering and Psychology in 2017. Some of his interests include competitive rock climbing and video gaming.

Ryan Clue

Ryan Clue is from Kent, Washington. He is currently attending Bradley University with the goal of graduating with a Bachelor of Science in Electrical & Computer Engineering in 2017. Some of his interests include DOTA2 and 3-D computer graphic design.

Contact Us

Please view the Github for more resources & feel free to email us if you have any questions