Efficient Robot Following for Enhanced Formation Control in Limited Computing Environments
- Resource Type
- Conference
- Authors
- Jin, Pin; Wang, Yuhan; Zou, Zhuo; Gan, Zhongxue; Liu, Lizheng
- Source
- 2023 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET) Artificial Intelligence in Engineering and Technology (IICAIET), 2023 IEEE International Conference on. :241-246 Sep, 2023
- Subject
- Bioengineering
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
General Topics for Engineers
Photonics and Electrooptics
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Transportation
Deep learning
PI control
Robot vision systems
Process control
Robot sensing systems
Light emitting diodes
Real-time systems
resource-constrained robots
robot following
serial tracking
PID control
- Language
This paper presents our work on achieving robot following using resource-constrained robots without relying on computationally expensive strategies like deep learning. Our system architecture is built upon the e-puck robot platform, which features various hardware components including a Time-of-Flight (ToF) sensor and LEDs. We propose two tracking algorithms: ToF tracking and LED tracking. The ToF tracking algorithm estimates the direction of the leader robot by oscillating the follower robot and collecting ToF readings from both sides. The LED tracking algorithm processes real-time camera frames, and utilizes a PID control mechanism to ensure that the follower robot maintains the desired position relative to the leader. By leveraging the unique capabilities of the e-puck platform, we enable resource-constrained robots to perform complex tasks such as tracking and following without high computational requirements. Our research contributes to the advancement of robotics in resource-constrained settings, opening up new possibilities for the deployment of intelligent and efficient robotic systems.