About CAIRSSThe Centre for AI Robotics in Space Sustainability
(CAIRSS, pronounced as “cares”) is a strategic initiative aligning with HKUST's
global research vision, addressing critical space sustainability issues through
interdisciplinary AI & Robotics research.
We focus on developing innovative solutions for space
sustainability challenges including space debris removal, in-space servicing,
assembly, manufacturing, and In-Situ Resource Utilization (ISRU). Leveraging
Hong Kong's unique position as a "super connector," we aim to
accelerate space innovation and commercialization.
§ Consisting of
o Development of advanced algorithms for
space sensing and perception, utilizing lidar, optical, and bio-signals.
o AI-based sensory data processing for
spacecraft GNC and decision making, as well as astronaut-assistive task
automation.
§ Pushing performance boundaries of onboard
autonomy under strict computational constraints.
§ Ensuring trustworthiness and reliability in
space environments.
§ Developing bio-inspired or AI-embodied mechanisms
for space locomotion, from surface and subsurface mobility solutions to
advanced manipulation technologies, and astronaut-assistive robotic functions.
§ Overcoming physical limitations in space
exploration.
§ Optimizing for low mass, volume, and power
requirements in space systems
RP1: Next generation of robotic manipulation for active
debris removal
This program develops a bio-inspired, AI-powered robotic manipulator
designed for the precise capture and removal of space debris in unstructured
orbital environments. Unlike conventional rigid-body robotic arms, our solution
integrates high-precision rigid joints with adaptive stiffness-controlled
actuation, enabling both effective targeting of fast-moving debris and
compliant, dexterous handling to prevent fragmentation upon contact. A hybrid
sensing system combining vision-based motion prediction with tactile feedback
ensures robust operation in dynamic conditions, while lightweight, self-healing
materials enhance durability against target object impacts. By balancing
machine learning for real-time decision-making with mechanically intelligent
design, this system addresses the critical gap between force-sensitive
manipulation and autonomous adaptability, key to sustainable orbital cleanup.
RP2: Bio-Inspired Robotic Penetrator for Lunar Polar
Regolith Resource Prospecting and Excavation
This program pioneers a novel bio-inspired
robotic penetrator to overcome the challenges of prospecting and excavating
water ice and volatile-rich regolith in the Moon’s polar regions. Drawing
inspiration from insect and/or earthworm
reciprocating/peristaltic motion and root growth mechanisms, the penetrator
combines self-burrowing locomotion with low-power, high-efficiency excavation
to minimize energy consumption in extreme lunar conditions. The system features
modular, deformable segments that adapt to subsurface density variations, while
embedded spectroscopic sensors enable real-time in-situ resource analysis. A
hybrid actuation system allows precise control over penetration depth and
sample extraction without excessive regolith disturbance. This approach
addresses critical limitations of conventional drilling, offering gentle yet
effective prospecting to support future In-Situ Resource Utilization (ISRU)
missions.
RP3: Assistive Robotics enhancing Astronaut Safety and
Productivity in Space Operations
This program focuses on intelligent
assistive robotics to enhance astronaut efficiency and safety during complex
space operations. Leveraging multi-modal sensing and AI-powered control, the
system integrates eye-tracking technology with haptic feedback interfaces to
enable astronauts to remotely operate robotic manipulators with unprecedented
precision and reduced cognitive load. The robotic platform features
context-aware autonomy, allowing it to interpret gaze-directed commands while
compensating for microgravity-induced motor control challenges. Advanced
computer vision algorithms map eye movements to robotic actions, while
predictive assistance anticipates astronaut intent to
streamline tasks like equipment maintenance or scientific sample handling. By
combining human-centric design with adaptive machine learning, this system
reduces operation time while minimizing errors during critical procedures,
fundamentally transforming human-robot collaboration in
space exploration.
Centre for AI Robotics in Space
Sustainability (CAIRSS)
Space Science & Technology Institute
(SSTI), Hong Kong University of Science & Technology
© 2025 All Rights Reserved