Object navigation requires an embodied agent to find an object in an unseen environment under partial observability and a limited motion budget. Existing methods primarily optimize where the robot should go next by ranking candidate destinations. In contrast to these methods, we present SOR-Nav, a hierarchical navigati...
Yuan Ji, Zi-Rui Li, Yu-Xin Cai et al.· 0 citations
Existing car-following prediction methods mainly optimize trajectory accuracy, while rarely considering whether predicted disturbances propagate realistically along a vehicle platoon. This limitation may lead to accurate but string-unstable predictions. We propose SSP-DMGTimeNet, a physics-constrained learning framewor...
Yu-Hang Wang, Kai-Lang Ma, Zirui Li et al.· 0 citations
DynaDreamer is proposed, a dynamics-augmented Dreamer-style reinforcement learning method to address the problem of egocentric driving by augmenting the WM with an explicit ego-dynamics prior, and improves task success rates over the strongest baseline.
Zhidong Wang, Jingsong Liang, Zirui Li et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.