It is shown that learning linear autoencoders and latent dynamics with multi-step prediction-reconstruction losses with backward losses provably recovers the structure needed for stabilization, and why both reconstruction and a multi-step horizon are essential.
Paul Lutkus, L. Lindemann, Stephen Tu· 0 citations
Safety-aware motion planning remains a challenge in robotics, especially when missions are time-critical and are under complex specifications. In this paper, we propose safety-aware-stl-mppi, a computationally efficient sampling-based receding-horizon planning framework designed to promote satisfaction of constraints e...
Yi-Qi Zhao, Taekyung Kim, Hideki Okamoto et al.· 0 citations
Two encodings of the problem of planning multi-agent paths that satisfy constraints written in STL-GO are presented, one based on mixed-integer programming (MIP) and another based on satisfiability modulo theory (SMT), with soundness guarantees.
This work synthesizes a prescribed performance controller by combining nominal model compensation with a learned state-dependent feedback gain and establishes two complementary guarantees: a semi-global conservative actuator-authority-based certificate for prescribed performance and input satisfaction, and a local data...
Panagiotis S. Trakas, Omid Mirzaeedodangeh, L. Lindemann· arXiv.org· 0 citations
A unified probabilistic perspective on CP and DRO is developed by viewing both as ways to turn finite calibration data into a data-dependent quantile estimator that a test score falls below with high probability.
Kehan Long, Yi-Qi Zhao, Pol Mestres et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.