The program, speakers and dates for CVPR 2027 are provisional and may change.
// CVPR 2027 Workshop · 5th edition
In conjunction with CVPR 2027
Computer vision and machine learning for spacecraft that plan their own trajectories, robots that explore together, and constellations that respond to a changing Earth.
Submission deadline · 21 Feb 2027
01 / Mars surface ops
Three rovers and a scout helicopter share one map, split the unexplored frontier between them and send the nearest rover to every promising sample site.
AI for space robotics and autonomous operations
02 / Autonomous GNC
The upper stage lights while still attached, the booster flips and returns to land, and the upper stage climbs from a parking orbit into its transfer orbit.
Vision and learning for guidance, navigation, and control
03 / Earth response
Satellites detect floods and wildfires onboard, relay the alert over inter-satellite links, re-task a neighbour to confirm and downlink it to the nearest ground station.
Distributed AI for mission operations and adaptive sensing
// 01 — Mission timeline
// 02 — Mission brief
The AI4Space workshop at CVPR 2027 brings together researchers, engineers, and space domain experts to address the unique challenges of deploying AI in space environments. As missions demand greater autonomy, resilience, and on-board intelligence, this workshop explores how computer vision and machine learning can enable breakthrough capabilities in spacecraft navigation, planetary exploration, satellite data analysis, and orbital operations.
AI4Space fosters collaboration between machine learning researchers and space scientists, focusing on topics such as vision-based navigation, onboard learning, sim-to-real transfer for robotics, multi-sensor fusion for Earth observation, and AI-driven mission planning—shaping the future of intelligent space exploration.
// 03 — Areas of focus
AI4Space 2027 invites work across the full pipeline, from onboard hardware to mission design. Two areas of particular growth get a spotlight this year:
For autonomous mission orchestration and the automation of space system and subsystem design, reflecting emerging trends and growing interest within the space community.
Integrating AI with new computing paradigms and sensing technologies, such as neuromorphic computing and sensing, to enable new capabilities and applications for future space missions.
Tool-using agents for systems engineering, multi-agent design workflows, requirements generation and traceability, automated design-space exploration, and agent-assisted verification and validation.
Neuromorphic computing, event-based sensing, and in-memory computing.
Multimodal reasoning across imagery, telemetry, geospatial metadata, and scientific measurements; pretraining on EO and astronomy datasets; and evaluation protocols.
Federated and distributed learning, tip-and-cue architectures, event-driven operations, and autonomous mission orchestration.
Rendezvous and docking, pose estimation, entry, descent, and landing (EDL), and space object tracking.
Perception, planning, control, multi-robot coordination, and learning-based dynamics modeling.
Planetary mapping, astronomical image analysis, anomaly detection, risk-aware decision-making for space exploration, and physics-informed modeling.
Quantization, pruning, neural architecture search, and parameter-efficient adaptation of foundation models.
AI accelerators, hardware–software co-design, and integration with flight computing systems.
Uncertainty estimation, out-of-distribution detection, explainability, fault diagnosis and recovery, and safety assurance.
Automated labeling, synthetic data generation, adaptation across sensors and missions, bridging the sim-to-real gap, and benchmark development.
// 04 — Keynotes
KEYNOTE 01Technical Fellow in AI and Co-head of the Artificial Intelligence Group, NASA Jet Propulsion Laboratory
Distributed Autonomy in Space Operations and Earth Observation
Dr. Steve Chien is a Technical Fellow in Artificial Intelligence and Co-head of the Artificial Intelligence Group at the Jet Propulsion Laboratory (JPL), California Institute of Technology, with decades of experience developing and deploying AI for space missions. His work on autonomous science, automated planning and scheduling, and intelligent spacecraft operations, including Earth Observing One, ESA's Rosetta Orbiter, and NASA's Mars 2020 mission, provides a strong perspective on the transition of AI from research to operational space systems and on enabling increasingly autonomous spacecraft.
KEYNOTE 02Assistant Professor, Politecnico di Milano (DART Lab)
Disruptive Computing and Sensing Paradigms for EdgeAI in Space
Dr. Paolo Panicucci is an Assistant Professor at Politecnico di Milano and a researcher at the Deep-space Astrodynamics Research and Technology Laboratory (DART Lab), specializing in computer vision, vision-based navigation, and spacecraft autonomy. His experience in autonomous GNC and projects such as EXTREMA and LUMIO brings expertise in AI-enabled perception and navigation, while his talk on ESA ION-V will introduce neuromorphic computing and sensing for Earth Observation.
KEYNOTE 03Chief Technology Officer, Redwire
Automating Satellite Manufacturing and Design
Al Tadros is the CTO of Redwire. With over 30 years of experience in aerospace engineering and leadership, he specializes in AI-enabled autonomy, machine vision, robotics, and rendezvous and proximity operations. Previously, he served as Chief Growth Officer and Executive Vice President of Space Infrastructure at Redwire and Vice President of Space Infrastructure and Civil Space at Maxar Technologies, contributing to 13 space missions and advancing in-space manufacturing, satellite servicing, and autonomous systems. He leads Redwire’s involvement in Stanford University’s Center for AEroSpace Autonomy Research, focusing on AI-driven autonomous reasoning in space. He has served on national space technology advisory boards, holds patents in attitude control systems, and contributed to early telerobotics research for the International Space Station.
KEYNOTE 04Associate Professor, Georgia Tech | Director, Planetary Robotics Lab
AI for Risk-Aware Decision Making in Space Robotics
Dr. Masahiro (Hiro) Ono is an Associate Professor at the School of Aerospace Engineering at Georgia Tech and Director of the Planetary Robotics Lab (PRL). Prior to joining Georgia Tech, he spent 13 years at NASA’s JPL, where he developed autonomous navigation algorithms for the Perseverance Mars rover, led the development of Exobiology Extant Life Surveyor, a snake-like robot designed to explore the potentially habitable subsurface ocean of Enceladus, and led multiple research initiatives in AI and robotics. He also served as Supervisor of JPL’s Robotic Surface Mobility Group and as a rover operator for Perseverance.
// 05 — Fishbowl session
Chief Technology Officer, Setanta Space
Professor, Stanford University | Director of Autonomous Vehicle Research, NVIDIA
Associate Professor, Stanford University | Founding Director, Space Rendezvous Laboratory
// 06 — Challenge
Join us for an exciting AI challenge featuring two cutting-edge streams focused on next-generation space perception!
// 07 — Support
// 08 — Comms
Questions? Ideas? Want to collaborate? Reach out to the AI4Space organizing team.
ai4space.organizers@gmail.com