onsite
Reinforcement Learning & Controls Research Scientist - Spot Behavior - Boston Dynamics
Research Engineer
Research scientist focused on designing, training, and deploying reinforcement‑learning policies that integrate with low‑level controllers to enhance Spot's locomotion, agility, and robustness on real‑world terrain.
About the role
Key Responsibilities
- Design, implement, and evaluate reinforcement‑learning algorithms for quadruped locomotion and mobility.
- Integrate RL policies with Spot's existing control stack, including PD/PID and whole‑body controllers.
- Develop simulation pipelines and transfer learning techniques to bridge the gap between simulation and real‑world deployment.
- Collaborate with hardware and software teams to tune low‑level controllers for stability and performance on challenging terrain.
- Analyze experimental data, iterate on model architectures, and document findings for internal knowledge sharing.
Requirements
- Ph.D. or equivalent experience in Reinforcement Learning, Robotics, Control Theory, or a related field.
- Strong programming skills in Python and C++, with experience in robotics middleware such as ROS.
- Hands‑on experience developing and deploying RL policies on real hardware, preferably legged robots.
- Solid understanding of low‑level control (PD/PID, whole‑body control) and ability to tune controllers for dynamic environments.
- Proven track record of publishing research or delivering production‑grade solutions in robotics or autonomous systems.
Skills
reinforcement learningpythonc