Your application is being processed
Learning-Enhanced Interactive decision making for Autonomous Driving
Job Number: P24INT-44
Honda Research Institute USA (HRI-US) is seeking a talented and motivated PhD-level research intern to explore the integration of learning-based methods with interaction-aware motion planning for autonomous driving. The intern will contribute to the development of hybrid learning-control frameworks for automated driving systems and micromobility platforms operating in complex real-world environments. This internship investigates how learning-based methods can enhance interactive planning by enabling autonomous vehicles to reason about and adapt to the behaviors of surrounding human agents, such as drivers and pedestrians. Emphasis will be placed on incorporating learned models of intent, response, and social compliance into motion planning algorithms to achieve safe, interpretable, and adaptive behavior in complex multi-agent environments.
Mountain View, CA
Key Responsibilities
|
|
- Conduct a comprehensive review of learning-based interactive planning techniques, including imitation learning, inverse reinforcement learning, and multi-agent reinforcement learning.
- Design, implement, and evaluate novel planning algorithms that incorporate predictive or learned models of human behavior in traffic environments.
- Develop hybrid frameworks combining model-based planning with learning-based components such as learned cost functions, behavioral priors, or policies.
- Integrate the developed algorithms into an existing motion planning stack and evaluate them in simulation with realistic driving and interaction scenarios.
- Collaborate closely with HRI scientists to test and refine the developed methods and support their deployment in internal research platforms.
- Contribute to academic publications targeting top conferences such as CoRL, ICRA, RSS, or ITSC
Minimum Qualifications
|
|
- Ph.D. candidate in Robotics, Computer Science, Artificial Intelligence, or a related field.
- Strong background in at least one of the following: interactive motion planning, hybrid AI, learning enhanced planning, behavior prediction, inverse reinforcement learning, imitation learning, multi-agent systems.
- Experience with C++ and Python.
- Familiarity with motion planning frameworks and simulation environments for autonomous vehicles.
- Self-motivated and able to conduct independent research and prototyping.
Bonus Qualifications
- Research experience in Robotics/Automated vehicles Motion Planning and Control, Machine learning.
- Prior published research on hybrid control, multi-agent interaction handling, behavior modeling
- Experience integrating learning models into control/planning pipelines (e.g., learned cost functions, behavior cloning policies)..
- Strong software development experience.
- Experience with using ROS-framework packages.
- Experience using public datasets and simulators for navigation research
Years of Work Experience Required |
0 |
Desired Start Date |
1/26/2026 |
Internship Duration |
3 Months |
Position Keywords |
Learning-enhanced planning, hybrid learning-control systems, multi-agent interaction, autonomous driving
|
|
|
|
|
Alternate Way to Apply
Send an e-mail to careers@honda-ri.com with the following:
- Subject line including the job number(s) you are applying for
- Recent CV
- A cover letter highlighting relevant background (Optional)
Please, do not contact our office to inquiry about your application status.