Additively Manufactured Open-Source Quadruped Robots for Multi-Robot SLAM Applications

Abstract

This thesis presents the design and development of the quadruped robot Squeaky, created as a research and learning platform for single and multi-robot simultaneous localization and mapping (SLAM), computer vision, and reinforcement learning. Affordable robots are increasingly essential for scaling from single-robot to multi-robot applications, as costs can rise exponentially as fleet size increases. SLAM is a critical feature for enabling a robot to perceive and localize itself within its environment, supporting applications such as cave exploration, disaster assistance, and remote inspection. To enhance efficiency, a fleet of robots can collaborate to merge individual maps for multi-robot SLAM. Squeaky is an affordable quadrupedal robot with adaptable hardware and software capable of creating merged maps from multiple robots over a shared network. It is also open-sourced for the benefit of the research and educational community. This work covers Squeaky's full design and development process, with validated results demonstrating the platform's effectiveness for research and educational applications.