Skip to Main Content
You are now in the main content area

Geometric GNC of Terrestrial and Space Robotics

At the ELIXIR Lab, we develop advanced geometric GNC frameworks that empower robots to operate autonomously and reliably in uncertain, dynamic environments. While our work is inspired by the rigorous demands of space robotics, all methodologies are intentionally designed to be immediately transferable to aerial and ground-based systems, enabling autonomy across sectors such as aerial manipulation, logistics, manufacturing, infrastructure inspection, and disaster response.

Our approach is grounded in Lie group theory, differential geometry, and nonlinear control, and is enriched with Deep Reinforcement Learning (DRL) to blend theoretical guarantees with real-time adaptability. Every research direction is numerically simulated and experimentally validated in both software and hardware-in-the-loop environments to ensure practical viability.

GNC for Moving-Base Manipulators in Proximity Operations

We investigate the modeling, perception, planning, and control of robotic manipulators mounted on dynamic platforms — such as drones, rovers, and satellites — engaged in tasks that involve interaction with noncooperative targets. Originating from orbital servicing applications, this research focuses on the development of singularity-free geometric modeling, safe and robust output-tracking GNC strategies, and the integration of DRL-based planners for real-time, adaptive behavior in complex settings.

We extend our GNC principles to design mission-level autonomy architectures for fleets of aerial drones, ground robots, and satellite servicers. These systems are developed to execute coordinated, multi-target operations with a high degree of autonomy and adaptability. Our research emphasizes resource-optimized mission planning, distributed collaborative localization, and priority-based scheduling—allowing robotic agents to dynamically allocate tasks, learn during idle periods, and execute complex, cooperative behaviors in real-world missions across air, land, and space.

Our work on autonomous rover systems directly targets the challenges of mobile robots operating in GPS-denied, off-road, agricultural, and search-and-rescue environments. We develop geometrically robust control techniques, traction optimization algorithms, and localization methods based on stochastic differential equations, ensuring reliable motion and decision-making across unstructured and deformable terrains.

Selected Publications

  1. M. Zarei and R. Chhabra, “Fault-Tolerant Multi-Modal Localization of Multi-Robots on Matrix Lie Groups, (external link) " IEEE Transactions on Robotics, under review.
  2. B. M. Moghaddam and R. Chhabra, “Safe Workspace Guidance, Navigation, and Target-Tracking Control of Space Manipulators," submitted to IEEE Transactions on Robotics, July 2025.
  3. B. M. Moghaddam and R. Chhabra, “Lagrange-Poincaré-Kepler Equations of Disturbed Space-Manipulator Systems in Orbit," submitted to Acta Astronautica, July 2025.
  4. M. R. Mottaghi, R. Chhabra, and W. Huang*, “Fast Traction Control of Planetary Rovers on Prescribed Trajectories with Wheel-Fighting Consideration,” AIAA Guidance, Control, and Dynamics, vol. 48, no. 6, pp. 1381-1396 , 2025.
  5. V. Gzenda and R. Chhabra, “Recursive Input-Output Linearization for Slow-Fast Realization of Nonholonomic Hamiltonian Control Systems," IEEE Transactions on Automatic Control, vol. 70, no. 6, pp. 3571-3586, 2025.
  6. V. Gzenda and R. Chhabra, “Affine Connection Approach to the Realization of Nonholonomic Constraints by Strong Friction Forces," Nonlinear Dynamics, vol. 112, no. 24, pp. 21627-21644, 2024.
  7. V. Gzenda and R. Chhabra, “Wheeled mobile robots on rough terrains as stochastic nonholonomic systems," IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp.1256-1262, 2024.
  8. M. Zarei and R. Chhabra, “Consistent Fusion of Correlated Pose Estimates on Matrix Lie Groups," IEEE Robotics and Automation Letters, vol. 9, no. 7, 2024.