Postdoctoral Research Scientist: Robotics and Control Systems

Honda Research Institute USA, Inc.; San Jose (CA); In loco; USD 100.000 - 130.000; Tempo pieno

Descrizione del lavoro

A leading research institute in robotics seeks a talented individual to develop and implement innovative robot behaviors and control algorithms. You will validate robot algorithms through simulations and data analysis, contributing to academic publications and patents. The ideal candidate holds a Ph.D. in a relevant field and possesses extensive hands-on experience in robotics and programming skills in C++ or Python. Strong communication skills are essential for this role, which also involves collaboration with global teams.

Competenze

• Ph.D. in mechanical engineering, robotics, electrical engineering, computer science, or a related field.
• Expert knowledge on robotics fundamentals like kinematics and dynamics.
• Extensive hands-on experience in planning, control, and recovery algorithms.

Mansioni

• Research and implement robot behaviors and interaction control algorithms.
• Develop simulation environments for data collection.
• Validate robot behavior algorithms through data analysis.

Conoscenze

Robotics fundamentals

Reinforcement learning

C++ programming

Python programming

ROS familiarity

Formazione

Ph.D. in mechanical engineering or related field

Strumenti

MuJoCo

Gazebo

PyBullet

Descrizione del lavoro

Key Responsibilities

• Research, ideate, and implement novel and robust model- and learning-based robot behaviors and interaction control algorithms, enabling contact-rich manipulation in various real-world application domains including dexterous object manipulation.
• Research, ideate, and implement robot reasoning algorithms for understanding and predicting failure, recovering from failure, and learning successful task strategies through adaptive intelligence that can enable generalizability to new scenarios.
• Develop and implement simulation environments for data collection.
• Validate dexterous robot behavior algorithms through comprehensive data analysis and successful demonstrations on hardware.
• Deliver results in accordance with the project timelines.
• Assist with compiling written and oral reports for executives within the company and performing demos.
• Contribute to the portfolio of patents, academic publications, and prototypes to demonstrate research value.
• Collaborate with our teams of scientists and engineers in Honda's regional and global R&D offices, as well as with our partners in academia and industry.
• Assist with supervising interns.

Minimum Qualifications

• Ph.D. in mechanical engineering, robotics, electrical engineering, computer science, or a related field.
• Expert knowledge on fundamentals of robotics. e.g. robot kinematics and dynamics, control systems, force control, haptics.
• Expert Knowledge on learning-based methods such as reinforcement learning, generative models and inference models.
• Extensive hands-on experience on development and successful implementation of planning, control and recovery algorithms for contact-rich manipulation in real-world.
• Proven experience with one or more physics simulators such as MuJoCo, Isaac Sim, PyBullet, Drake, or Gazebo.
• Familiarity with large models including VLMs, LLMs, and multi-modal foundation models.
• Strong track record of publishing in top robotics venues (e.g T-RO, RA-L, T-M, ICRA, IROS, RSS, CoRL, …)
• Solid programming skills in C++ or Python and strong familiarity with ROS.
• Self-motivated to advance the project and deliver on-time.
• Strong written and oral communication skills including development and delivery of presentations, papers, and technical documents.

Bonus Qualifications

• Strong familiarity with both model and learning-based approaches for manipulation.
• Extensive experience with algorithm and software development (and debugging) for complex robotic systems involving design, implementation and integration of multiple planners, controllers, and perception systems.
• Research experience and expertise in planning and control for dexterous manipulation with tactile robotic multi-fingered hands.