Research Intern: Motion Planning and Decision Making Under Noise and Prediction Uncertainty for Autonomous Driving - Honda Research Institute USA

Research Intern: Motion Planning and Decision Making Under Noise and Prediction Uncertainty for Autonomous Driving

Your application is being processed

Research Intern: Motion Planning and Decision Making Under Noise and Prediction Uncertainty for Autonomous Driving

Job Number: P26INT-02
Honda Research Institute USA (HRI-US) is seeking a self-motivated research intern to focus on interaction-aware motion planning and decision-making under perception noise, state uncertainty, and prediction uncertainty for autonomous driving (AD) applications. Autonomous vehicles frequently interact with pedestrians and other vehicles whose motions and intentions are uncertain and may have multiple possible outcomes. These interactions require autonomous vehicles to infer intentions in real time and adapt their motion plans accordingly. To ensure safe and reliable operation, the vehicle's movements must align with human expectations. This internship involves developing an interaction-aware motion planner for uncertain, dynamic environments within a model predictive control (MPC) framework. The research will integrate social norms (e.g., yielding to pedestrians at crossings or signaling intentions) into the planning process. The intern will collaborate closely with HRI scientists and engineers and have the opportunity to present research findings at leading academic conferences or publish them in journals.
Mountain View, CA

 

Key Responsibilities

 
  • Develop, implement, and test novel algorithms for autonomous driving, focusing on uncertainty- and interaction-aware motion planning to address perception noise, state uncertainty, and multimodal motion prediction uncertainty in dynamic environments involving multiple vehicles and pedestrians, within a model predictive control (MPC) framework.
  • Extend the MPC-based motion planning stack to incorporate uncertainty modeling (state, model, multimodal prediction), socially acceptable interaction-aware behaviors, real-time interactions, partial observability and occlusion.
  • Evaluate the performance, safety, and robustness of the proposed algorithms in simulation environments, including uncertain dynamic scenarios involving pedestrians and other vehicles.
  • Collaborate with HRI scientists and engineers to integrate findings into practical applications.
  • Submit research findings to leading academic conferences or journals.

 

Minimum Qualifications

 
  • Ph.D. candidate in Robotics, Computer Science, Mechanical Engineering, or similar fields.
  • Strong background and experience in at least one of the following: motion planning and decision-making under uncertainty, interaction-aware motion planning, occlusion-aware planning, active learning, or reinforcement learning.
  • Experience with model predictive control (MPC).
  • Programming and software design skills in C++ and Python.
  • Self-motivated and able to work independently.

 

Bonus Qualifications​

  • N/A

 

Years of Work Experience Required   0
Desired Start Date  May 2027
Internship Duration  3 Months
Position Keywords  

​Uncertainty-aware, Interaction-aware, Motion planning, Model Predictive Control (MPC),

Multi-modal motion prediction uncertainty, 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.