Our quadruped robot development project focuses on developing autonomous locomotion systems that combine cutting-edge hardware design with advanced reinforcement learning algorithms. We are working on custom hardware development and embedded systems to create innovative quadruped solutions.
Our hardware development focuses on creating a custom quadruped robot that combines innovative mechanical design with advanced embedded systems.
Custom chassis, leg mechanisms, and actuator systems optimized for agility and stability
Real-time control systems, sensor integration, and communication protocols
Custom leg mechanism design
RL training in simulation environment
Our reinforcement learning approach focuses on developing robust locomotion policies that can adapt to various terrains and conditions.
High-fidelity physics simulation for safe and efficient training
Algorithms that adapt to different surfaces and environmental conditions
Sim-to-real techniques for seamless deployment on physical hardware
We utilize the Deep Robotics Lite3 as our primary testing platform for validating RL algorithms and conducting real-world experiments. This platform allows us to bridge the sim-to-real gap effectively.
Interested in contributing to cutting-edge quadruped robot development? Join our team and gain valuable experience in robotics.
Applications are reviewed on a rolling basis. We encourage early submissions to secure your preferred position.