Figure is an AI Robotics company developing a general purpose humanoid. Our Humanoid is designed for corporate tasks targeting labor shortages and jobs that are undesirable or unsafe. We are based in Sunnyvale, CA and require 5 days/week in-office collaboration. It’s time to build.
We are looking for a Staff Whole-Body Manipulation Engineer. In summary, you will be developing new control software for our Humanoid robot, working both in simulation and on hardware. You will have the opportunity to develop new manipulation capabilities on the robot. We aim to develop a robot that can interact with the world in a stable, precise and dexterous manner, utilizing its whole-body for manipulation.
Responsibilities:
- Develop advanced manipulation algorithms: Design and implement whole-body control, multi-contact control and compliant control strategies to enable robust object manipulation without compromising robot stability. Ideal manipulation strategies:
- Enable precise control of forces and torques at the robot’s hands to achieve delicate and robust manipulation tasks.
- Leverage the robot’s entire body to maximize reachability, stability, and dexterity during object interactions.
- Don’t compromise balance and stability during dynamic object interactions.
- Work closely with perception and motion planning teams to seamlessly integrate sensor data and trajectory planning into manipulation tasks.
- Use simulation to develop and test new algorithms. Work with the simulation team to improve simulation fidelity.
- Rapidly iterate on hardware to validate new control algorithms, working with other teams to ensure performance requirements are met.
- Conduct System Diagnostics: Monitor and debug control performance to ensure stability, responsiveness, and safety during real-world operations.
- Research State-of-the-Art Techniques: Stay up-to-date with advancements in control theory, manipulation strategies, and robotics technology to keep Figure at the forefront of humanoid robotics.
Requirements:
- 4+ years of experience implementing optimization-based multi-DOF control algorithms (inverse dynamics, trajectory-optimization, model-predictive control, whole-body control)
- Understanding of spatial kinematics, rigid body dynamics, and related mathematical concepts
- Fluency in modern C++
- A proven track record implementing and deploying manipulation algorithms in simulation and on hardware
- Prior experience working on multi-DOF robot hardware
Bonus Qualifications:
- Experience with C++20 and Python
- Experience with multiple parts of the robotics software stack (modeling, estimation, planning, control)
