Research Intern: Multi-Robot Task Planning Using LLMs

Research Intern: Multi-Robot Task Planning Using LLMs - Honda Research Institute USA Job Number: P25INT-37 Honda Research Institute USA (HRI-US) is seeking a self-motivated Research Intern to join our Cooperative-Intelligence Mobility Research (CMR) group. This internship offers the unique opportunity to work at the intersection of task planning, heterogeneous multi-robot systems, and cutting-edge foundation models, including Large Language Models (LLMs) and Vision-Language Models (VLMs). The intern will help advance next-generation autonomy and coordination frameworks for intelligent robot teams in dynamic environments. Mountain View, CA Key Responsibilities Conduct literature reviews to survey state-of-the-art methods in multi-robot task planning and foundation model integration. Devise novel solution strategies for long-horizon task planning in heterogeneous multi-robot teams using LLMs and VLMs, including failure detection and replanning. Implement proof-of-concept systems and integrate them with simulation or robotic platforms. Design and execute evaluation experiments to benchmark performance and generalization. Contribute to peer-reviewed publications in leading robotics and AI journals and conferences. Minimum Qualifications Currently enrolled in a PhD program in Robotics, Computer Science, Electrical Engineering, or a related field. Demonstrated experience with Vision-Language Models (e.g., CLIP, BLIP, Flamingo, etc.). Solid background in robotics and multi-agent systems. Strong programming skills in Python and familiarity with ML libraries (e.g., PyTorch, TensorFlow). Excellent communication and documentation skills. Bonus Qualifications Prior experience integrating LLMs (e.g., GPT, PaLM, LlaMA, Qwen) with perception or planning systems. Hands‑on experience with ROS, Gazebo, or other robotics simulation frameworks. Hands‑on experience with implementation and testing algorithms on physical platforms/robots. Familiarity with transformer‑based architectures for multimodal learning. Record of peer-reviewed publications in robotics, AI, or computer vision conferences. Years of Work Experience Required: 0 Desired Start Date: 8/31/2026 Internship Duration: 3 Months Position Keywords: LLMs, VLMs, Symbolic Planning, Scene Understanding, PDDL, Task Planning, Long-Horizon Planning #J-18808-Ljbffr Honda Research Institute USA