Research

We aim to develop AI techniques to confront the fundamental challenges in the optimization of networked autonomous systems, such as robotic systems and intelligent edge networking devices, especially in the presence of uncertainty.

Two research thrusts will be pursued:

  1. Trustworthy AI decision under uncertainty for autonomy: This thrust will tackle reliability challenges for autonomy under uncertainty. This is achieved by learning accurate 3D world models from LiDAR and camera data to serve as environment emulators, which enable the learning of certifiably safe actions based on data-driven probabilistic reachable sets of stochastic dynamic systems.
  2. AI-driven optimization for distributed autonomy on the edge: This thrust will achieve scale in autonomy and tackle practical AI deployment challenges on the edge. It develops distributed optimization algorithms over communication graphs to solve multi-agent problems, including motion planning, reinforcement learning (RL), and stochastic games. We also propose efficient distributed training strategies employing coding theory for reliable learning and optimization on the edge. These techniques enable peer-to-peer communication, local storage, and computation for multi-robot systems, and enhance networked edge devices with intelligent resource management.

Publication

  1. P. Sun, N. Singh, A. Dutta, J. Xie, "Benchmarking Small LLMs at the Edge for UAV Command Synthesis and Path Planning," in Proceedings of AIAA Aviation, San Diego, CA, June 2026.
  2. A. Dutta, N. Singh, P. Sun, J. Xie, "Onboard Voice-to-Mission UAV Autonomy with Language-Conditioned Navigation and Semantic Perception," in Proceedings of AIAA Aviation, San Diego, CA, June 2026.
  3. S. Kim, W. Chung, Z. Dai, D. Bhatt, A. Shukla, H. Su, Y. Tian and N. Atanasov, "Seeing the Bigger Picture: 3D Latent Mapping for Mobile Manipulation Policy Learning," in Proceedings of IEEE International Conference on Robotics and Automation (ICRA), Vienna, Austria, June 2026.
  4. S. Kim, B. Pak, K. Long, Y. Tian and N. Atanasov, "4D Latent Mapping for Mobile Manipulation Policy Learning," Workshop on Multi-Modal Spatial AI for Robust Navigation and Open-World Understanding and Workshop on Semantics for Reliable Robot Autonomy at IEEE ICRA, Vienna, Austria, June 2026.
  5. S. Liu, N. Atanasov and S. Koga, "MATT-Diff: Multimodal Active Target Tracking by Diffusion Policy," in Proceedings of Learning for Dynamics and Control (L4DC), June 2026.
  6. A. Inbaraj, J. Chen, "Physics-Informed Neural Models for Uncertain Trajectory Prediction in Urban Air Mobility," Journal of Aerospace Information Systems, Vol. 23, No. 4, 2026.
  7. L. Garcia, J. Xie, "Formation-Guided Hierarchical Reinforcement Learning for UAV Swarm-Assisted MEC," accepted by DCOSS-IoT, 2026 (Poster).
  8. J. Xiang, X. Chen, J. Chen, "Intention-Guided Airport Surface Trajectory Forecasting Using a Vision Language Model," in Proceedings of AIAA SciTech 2026, Orlando, FL, Jan. 2026.
  9. Xiang, J., & Chen, J. (2025). Data-driven probabilistic trajectory learning with high temporal resolution in terminal airspace. Journal of Aerospace Information Systems, 1–11.
  10. Ermanno Bartoli, Rebecca Stower, Bryan Donyanavard, Hanna Werner, Jana Tumova, Iolanda Leite. The Need for (Robot) Speed: Offloading Heavy Computations Improves Response Time and User Experience in Spoken Interactions. 2025 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN). (Accepted)
  11. E. Sebastián, T. Duong, N. Atanasov, E. Montijano and C. Sagüés, "Physics-Informed Multi-Agent Reinforcement Learning for Distributed Multi-Robot Problems," IEEE Transactions on Robotics (T-RO), 2025 (Accepted).