Gradient based reachability analysis for goal-oriented guidance. Application to Eros proximity operations
The recent growing interest in small solar system bodies such as asteroids and comets for scientific inspection, exploitation of resources, and planetary defense reasons is pushing the development of innovative engineering solutions to better investigate these celestial bodies. Current approaches for autonomous proximity operations often relies on tracking a reference trajectory previously designed and optimized on-ground. An innovative concept developed in recent years is proposing a paradigm shift towards autonomous goal-oriented strategies, providing the probe with high-level tasks and enabling replanning capability on-board. A gradient base reachability exploration scheme is introduced in this work to efficiently compute high reward paths with limited computational resource. The methodology is tested in a case study involving proximity operations to asteroid (433) Eros, comparing the proposed approach with other exploration techniques and showing promising results.
Code (0)
등록된 구현이 없습니다.
Methods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
NavArena: Automated Construction of Goal-Oriented Navigation Benchmarks from 3D Gaussian Splatting Reconstructions
Fixed 3D Gaussian Splatting (3DGS) reconstructions provide realistic novel views but lack the traversability constraints, valid goals, and closed-loop protocols required for navigation evaluation. We introduce NavArena, …
Visual NavigationDual Advantage Fields
Offline goal-conditioned reinforcement learning requires both long-horizon reachability estimates and local action comparisons. Dual goal representations provide value fields that capture global goal reachability, but th…
Reinforcement LearningCoupled Input-Output Dimension Reduction: Application to Goal-oriented Bayesian Experimental Design and Global Sensitivity Analysis
We introduce a new method to jointly reduce the dimension of the input and output space of a function between high-dimensional spaces. Choosing a reduced input subspace influences which output subspace is relevant and vi…
Combinatorial OptimizationDiagnosticDimensionality ReductionExperimental Design+1Goal Space Abstraction in Hierarchical Reinforcement Learning via Set-Based Reachability Analysis
Open-ended learning benefits immensely from the use of symbolic methods for goal representation as they offer ways to structure knowledge for efficient and transferable learning. However, the existing Hierarchical Reinfo…
Hierarchical Reinforcement Learningreinforcement-learningWalk the Random Walk: Learning to Discover and Reach Goals Without Supervision
Learning a diverse set of skills by interacting with an environment without any external supervision is an important challenge. In particular, obtaining a goal-conditioned agent that can reach any given state is useful i…
continuous-controlContinuous Control