Forward reachability of aerospace vehicles is critical for hazard control and risk mitigation. However, the governing equation and its data assimilation are two of the central questions to predict the reachable geometry and probability. We propose a strategy for the geometric and probabilistic forward reachable domain using inverse optimization by deriving the stochastic governing equation in Langevin dynamics and assimilating the sparse data observations in virtual exploration. The optimization problem is expressed as an explicit functional of the stochastic governing equation and an entropy criterion of stochastic complexity. Lévy–Khinchin–Itô decomposition, pseudotime stepping, pseudoenergy functional, and stochastic trajectory exploration are developed for the jump discontinuity of control inputs, the computational speedup of algorithmic convergence, the best initialization of data assimilation, and the prognostics of safety-critical factors, respectively. Numerical results show that the safety measure on the forward reachable domain is robust to avoid hazards and mitigate risks. In conclusion, this new strategy can determine the evolving probability and the propagating geometry of the dynamic forward reachable domain.
Geometric and Probabilistic Forward Reachable Domain of Aerospace Vehicles
IEEE Transactions on Aerospace and Electronic Systems ; 59 , 6 ; 9527-9538
2023-12-01
4042982 byte
Article (Journal)
Electronic Resource
English
Fast calculation method for helicopter reachable domain in forward flight
British Library Online Contents | 2014
|Evaluation of the Reachable Domain of Launch Vehicles Under Thrust Fault
Springer Verlag | 2025
|Reachable domain-based blockade orbit optimization method
European Patent Office | 2024
|