Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2309.16698

Autonomous Guidance Navigation and Control of the VISORS Formation-Flying Mission

Tommaso Guffanti·Toby Bell·Samuel Y. W. Low·Mason Murray-Cooper·Simone D'Amico
View original source ↗Read source document ↗Export metadata ↓

Abstract

Virtual Super-resolution Optics with Reconfigurable Swarms (VISORS) is a distributed telescope mission for high-resolution imaging of the Sun using two 6U CubeSats flying in formation in a Sun-synchronous low-Earth orbit. An optics spacecraft carries a photon sieve acting as a high-resolution lens in the extreme ultraviolet spectrum, while the image passing through the sieve is focused on a detector spacecraft. This paper presents the newly conceived design of the on-board guidance, navigation and control (GNC) system, which is highly autonomous, robust, passively safe, and validated under realistic mission simulations. The primary objective of the GNC system is to establish a passively safe and high-precision formation alignment at 40-meter separation, with sub-centimeter relative navigation and position control accuracy, over repeated observations of 10-second duration. Science mission success rates are assessed via Monte-Carlo analyses under realistically modelled uncertainties stemming from sensing errors, maneuver errors, unmodelled dynamics, and erroneous knowledge of internal spacecraft components. Precise real-time relative navigation is achieved by carrier phase differential GPS with integer ambiguity resolution. Precise control over short baselines is achieved via closed-loop optimization-based stochastic model predictive control with centimeter-level accuracy. Control at far range and during approach is achieved by closed-form impulsive control with meter-level accuracy. Passive safety is enforced throughout the mission to mitigate collision risks even under critical subsystem failure. Beyond VISORS, this work also realizes the crucial insight that the described GNC architecture is generalizable to other distributed space missions where accuracy and fault-tolerant safety are key requirements, such as rendezvous, proximity operations, and swarming missions.

Explore related subjects

cs.RO

Keep this discovery

BibTeXRIS

Tommaso Guffanti, Toby Bell, Samuel Y. W. Low, Mason Murray-Cooper, Simone D'Amico. 2023-08-12. Autonomous Guidance Navigation and Control of the VISORS Formation-Flying Mission. https://arxiv.org/abs/2309.16698

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies

Vision-language-action (VLA) policies leverage pretrained vision-language backbones to achieve strong cross-task generalization. A leading design couples this backbone with a dedicated continuous action head trained via diffusion or flow matching. However, such heads rely on iterative multi-step sampling, for example 10 Euler steps in $\pi_{0.5}$. This creates an inference bottleneck that produces stop-and-go movement in the robot and slower task completion. We introduce IMLE-VLA, which replaces the iterative action head with a single-step conditional generator trained via conditional Implicit Maximum Likelihood Estimation (cIMLE). The cIMLE objective promotes multimodal action coverage, avoiding the mode collapse of naive regression heads while eliminating multi-step sampling entirely. When IMLE-VLA is applied to $\pi_{0.5}$, it increases inference frequency 3.67x (55 Hz vs. 15 Hz), enabling up to 11x higher action throughput. On the 40-task LIBERO benchmark, IMLE-VLA achieves the highest average success rate (98.0%) among all baselines while leading in inference frequency. Under the test-time perturbations of LIBERO-plus, IMLE-VLA retains $\pi_{0.5}$'s robustness while other baselines degrade sharply, confirming that the cIMLE head preserves generalization. Real-world experiments on a Franka Emika Panda across four tasks demonstrate smoother motion (2.2x to 3.0x lower jerk) and faster task completion, with IMLE-VLA outperforming $\pi_{0.5}$ on every task and reducing average VLA inference time per episode by 3.9x to 6.6x. Videos and code are available at https://kianhk6.github.io/IMLE-VLA/

cs.RO↗
arXiv2026

ObstaDiff: Generalizable Diffusion Policy Learning via Obstacle-aware Representations

Imitation learning has achieved impressive results in robotic manipulation, yet most existing approaches assume clean backgrounds and lack explicit mechanisms for obstacle-aware motion generation. Extending such policies to cluttered, real-world scenes with unstructured obstacles remains a key generalization challenge. We present ObstaDiff, a decomposed diffusion-policy framework with a lightweight obstacle-aware visual encoder. ObstaDiff extracts a structured target-obstacle-background representation, enabling the downstream alignment policy to generate end-effector trajectories toward a target-centered bottleneck pose while reasoning about surrounding obstacles. We evaluate ObstaDiff on 61 real-robot greenhouse trials per method (366 executions in total). ObstaDiff achieves 75.41% average task success and 8.20% average obstacle collision rate, outperforming representative imitation-learning baselines and improving generalization in cluttered agricultural scenes.

cs.RO↗
arXiv2026

Testing Between the Test Cases: Proving End-to-End Steering in Conditions You Never Drove

AI-based automated vehicle testing is challenging because a model that passes every test condition can still fail in the real world. Formal verification offers a way to directly address this gap. On a simulated highway and an arterial road we trained two small end-to-end steering networks each in CARLA, one on clear conditions alone and one on clear, fog, night and low sun. All four models were driven against a 2.19 ft lane-departure budget. Without driving again, we used bound propagation, a formal method that reads the trained weights, to compute how far steering can drift at every disturbance strength between two captured images. One calculation covers more than a campaign could drive: on the arterial it spans 133 poses, where ten intensities each would be 10^133 combinations, in minutes on one GPU. Not only did formal verification find conditions that broke the clear-trained policy without simulation testing, it provided some preliminary evidence for potential failures between the test cases. Our overall conclusion is that formal verification is a viable complement to simulation, and could be adopted as a part of verification and validation for automated driving.

cs.RO↗