At Toyota Research Institute (TRI), we’re on a mission to improve the quality of human life. We’re developing new tools and capabilities to amplify the human experience. To lead this transformative shift in mobility, we’ve built a world-class team advancing the state of the art in AI, robotics, driving, and material sciences. The Opportunity
Our team at Toyota Research Institute is looking for a motivated and capable postdoc to conduct research on the computational and machine learning aspects of human-robot interaction. Areas of interest include applying foundation models to human-robot interaction, with an emphasis on understanding human-robot collaboration and how it evolves as foundation models become increasingly central to robotic manipulation. We’re looking for a postdoc with a solution driven attitude who can partner closely with our research team and turn ideas into impactful results. Relevant background topics include robot foundation models (e.g., VLAs, WAMs), imitation/reinforcement learning, leveraging language conditioning in trajectory planning/prediction, and decision-making under uncertainty, as well as human-machine teaming/shared control. The Challenge
In this project, we’re exploring novel approaches to efficiently learn how humans and robots collaborate. Our overall goal is to create AI approaches that challenge more traditional approaches to novel interactions with a human collaborator and enable the development of new embodied interactive policies. The project will be focused on researching a new approach in a specific sub-area of the overall goal, towards publications in top-tier conferences, well-matched for a postdoc that is looking for an additional year of research before beginning a faculty position, or a more experienced existing faculty member looking to take a year-long sabbatical with us. The fellowship will be in our Cambridge, MA research center, where you will be working with researchers on meaningful research problems, with access to the resources needed to pursue ambitious projects and publications. Applicants with relevant publications in these fields are highly encouraged to apply.