Reinforcement Learning for Spacecraft Guidance
A robust guidance scheme for far-range rendezvous using RL with safety and observability considerations.
Learn more →Accepting PhD Students · 2026
Space Sustainability
& Autonomy Lab
Space for good
Developing intelligent, safe, and sustainable autonomous systems for space exploration, debris removal, and proximity operations at the University of Illinois Urbana-Champaign.
F
Forbes 30 Under 30Transportation & Aerospace · 2026
About the Lab
Solving the space debris problem through autonomy
SSA Lab is an academic lab in the Grainger School of Engineering at UIUC. We focus primarily on solving the space debris problem by addressing critical challenges in missions with rendezvous and proximity operations.
Our work spans the entire lifecycle of in-orbit servicing missions — from preliminary trajectory design to real-time autonomous guidance during the final approach. We combine classical astrodynamics with modern AI techniques including reinforcement learning, control barrier functions, and convex optimization.
We collaborate with partners across academia and industry, including MIT, the Australian Centre for Field Robotics, Astroscale, and JPL, to push the boundaries of what's achievable in autonomous space operations.
Active Debris Removal
Real-time detection, optimal trajectory planning, multi-target missions, adaptive guidance, multi-strategy ADR for safe capture and disposal.
On-Orbit Servicing
Real-time detection, trajectory planning, multi-target handling, adaptive guidance, safe capture and repair, refuel, and reuse.
Space Sustainability
Sustainable spacecraft design principles prioritizing end-of-life reuse and reducing the long-term debris footprint of space operations.
Latest Updates
News
Feb 2026
New Preprint: Meta-RL for Robust CBFs in Spacecraft Proximity Operations
A framework learning the full hierarchy of class-K functions in ICCBFs via meta-RL for safer, more fuel-efficient RPO.
arXiv 2602.07335Jan 2026
New Preprint: Learning Safety-Guaranteed, Non-Greedy CBFs Using RL
Two-stage RL framework addressing conservatism and recoverability of ICCBFs for safety-critical spacecraft control.
arXiv 2602.00366Dec 2025
Forbes 30 Under 30 — Transportation & Aerospace
Minduli named to the Forbes 30 Under 30 list for 2026 for pioneering spacecraft autonomy and debris removal technology.
Award2026
Joining UIUC as Assistant Professor
Founding the Space Sustainability and Autonomy Lab in the Department of Aerospace Engineering at UIUC.
Appointment2025
Visiting Scientist at MIT
Collaborating with Prof. Richard Linares on advanced autonomy and safety-critical guidance for spacecraft.
Collaboration2025
RL Guidance Paper Published
Our paper on robust trajectory design using reinforcement learning with safety and observability considerations published in Aerospace Science & Technology.
Publication2024–25
GTOC Podium Finishes
As part of TheAntipodes: 3rd place in GTOC11, 5th in GTOC12, and 3rd in GTOC13.
Competition2025
CORTEX & GRASP Frameworks
Two frameworks for autonomous in-orbit servicing accepted for conference presentation.
PublicationOur Team
Researchers
Minduli Wijayatunga
Director / Incoming Assistant Professor · Forbes 30 Under 30
- Incoming Assistant Professor of Aerospace, UIUC — 2026
- Adjunct Assistant Professor of Aerospace, UIUC — 2025–2026
- Visiting Scientist, MIT — 2025–2026
- Research Associate, Australian Centre for Field Robotics — 2024–2025
Interests: Rendezvous & Proximity Operations, Astrodynamics, Reinforcement Learning, Control Barrier Functions, Multi-Objective Trajectory Optimization, Model Predictive Control, Convex & Indirect Optimization
What We Work On
Research Projects
PhD Work 6 projects
Improving Indirect Optimization Convergence
Scaling factors to connect energy-optimal to time-optimal and fuel-optimal problems for better convergence.
Learn more →
Preliminary Mission Design Tool (PMDT)
Generates multi-target, fuel- and time-optimal tours using low-thrust propulsion with J2, drag, eclipses, and duty cycles.
Learn more →
Global Trajectory Optimization Competitions
Team Antipodes — indirect and convex-based tools. GTOC11: 3rd, GTOC12: 5th, GTOC13: 3rd.
Model Predictive Control for Spacecraft Guidance
Convex-based MPC for low-thrust transfer trajectory guidance with real-time performance.
Ion Beam Deflection for Asteroids
A framework assessing the feasibility of deflecting hazardous asteroids via ion beams for planetary defense.
PostDoc Work 2 projects
CORTEX: Convex Optimization for Rendezvous
A sunlight-aware, robust convex-tracking scheme for final approach in RPO missions.
GRASP: Guidance for Autonomous Servicing
End-to-end transfer and guidance framework for spacecraft servicing missions.
Research Output
Publications
Selected journal and conference papers.
Preprints
- Meta-Reinforcement Learning for Robust and Non-greedy Control Barrier Functions in Spacecraft Proximity OperationsNew
- Learning Safety-Guaranteed, Non-Greedy Control Barrier Functions Using Reinforcement LearningNew
Journal Articles
- Robust trajectory design and guidance for far-range rendezvous using reinforcement learning with safety and observability considerations
- GTOC12: Results from TheAntipodes
- Convex-Optimization-Based Model Predictive Control for Space Debris Removal Mission Guidance
- Exploiting Scaling Constants to Facilitate the Convergence of Indirect Trajectory Optimization Methods
- Design and guidance of a multi-active debris removal mission
- Team theAntipodes: Solution Methodology for GTOC11
- Sustainability within Aotearoa New Zealand’s aerospace sector: current state and implications for the future
Conference Papers
- GRASP: An Integrated Framework for Mission Design and Guidance for Autonomous In-Orbit Servicing Missions
- System Design and Hardware-In-The-Loop Testbed Development for an Australian ISAM Demonstrator Mission
- CORTEX: Real-Time Trajectory Optimization and Guidance for the Final Approach Phase of In-Orbit Servicing Missions
- Assessing the Feasibility of Ion Beam-Based Asteroid Deflection for Planetary Defense
- State-dependent trust region for successive convex optimization of spacecraft trajectories
Join Us
Open Positions
I am currently looking for two PhD students to start in 2026. Please contact me if interested.
PhD: Rendezvous & Proximity Operations and Hardware Testing
Research focused on RPO algorithms and hardware testing for in-orbit servicing and debris removal missions.
Requirements
Astrodynamics background, previous publications, strong grades, some hardware experience.
PhD: Space Debris Removal
Developing autonomous guidance and mission design tools for active debris removal campaigns.
Requirements
Astrodynamics background, previous publications, strong grades, interest in debris removal.
Get in Touch