Auto-Platoon : Freight by example
Tharun V. Puthanveettil, Abhijay Singh, Yashveer Jain, Vinay Bukka, Sameer Arjun S
TL;DR
This work tackles multi-robot freight platooning by integrating object detection, state estimation, and monocular depth estimation to derive the leader's pose and generate follower trajectories. It combines a bio-inspired, software-latching leader–follower framework with a perception stack (YOLOv7 for detection and MiDaS for depth) and a dynamic planner that leverages IMU and depth data, supported by cooperative sensing and communication. The approach includes a Kalman-filter–based tracker with feature-based data association, metric-depth calibration, and a Stop-and-Follow planning strategy to maintain formation in dynamic environments. The system is implemented on a pair of low-cost robots with a high-performance off-board compute setup, demonstrating robust detection, tracking stability, depth estimation, and responsive planning, with clear pathways for edge-device deployment and enhanced sensing in future work.
Abstract
The work introduces a bio-inspired leader-follower system based on an innovative mechanism proposed as software latching that aims to improve collaboration and coordination between a leader agent and the associated autonomous followers. The system utilizes software latching to establish real-time communication and synchronization between the leader and followers. A layered architecture is proposed, encompassing perception, decision-making, and control modules. Challenges such as uncertainty, dynamic environments, and communication latency are addressed using Deep learning and real-time data processing pipelines. The follower robot is equipped with sensors and communication modules that enable it to track and trace the agent of interest or avoid obstacles. The followers track the leader and dynamically avoid obstacles while maintaining a safe distance from it. The experimental results demonstrate the proposed system's effectiveness, making it a promising solution for achieving success in tasks that demand multi-robot systems capable of navigating complex dynamic environments.