Recent work in Bayesian Experiment Design (BED) has shown the value of using Deep Learning (DL) to obtain highly efficient adaptive experiment designs. In this paper, we argue that a central bottleneck of DL training for BED is belief explosion. Specifically, as an agent progresses deeper into an experiment, the effective number of realisable beliefs grows enormously, placing significant sampling burdens on offline training schemes in an effort to gather experience from all regions of belief space. We argue that choosing an appropriate inductive bias for actor/critic networks is a critical component in mitigating the effects of belief explosion and has so far been overlooked in the BED literature. We show how Graph Neural Networks are particularly well-suited for BED DL training due to their domain permutation equivariance properties, resulting in multiple orders of magnitude improvement to sample efficiency compared to naive parameterizations.

Robotic solutions for quick disaster response are essential to ensure minimal loss of life, especially when the search area is too dangerous or too vast for human rescuers. We model this problem as an asynchronous multi-agent active-search task where each robot aims to efficiently seek objects of interest (OOIs) in an unknown environment. This formulation addresses the requirement that search missions should focus on quick recovery of OO1s rather than full coverage of the search region. Previous approaches fail to accurately model sensing uncertainty, account for occlusions due to foliage or terrain, or consider the requirement for heterogeneous search teams and robustness to hardware and communication failures. We present the Generalized Uncertainty-aware Thompson Sampling (GUTS) algorithm, which addresses these issues and is suitable for deployment on heterogeneous multi-robot systems for active search in large unstructured environments. We show through simulation experiments that GUTS consistently outperforms existing methods such as parallelized Thompson Sampling and exhaustive search, recovering all OOIs in 80% of all runs. In contrast, existing approaches recover all OOIs in less than 40% of all runs. We conduct field tests using our multirobot system in an unstructured environment with a search area of ≈75,000 m2. Our system demonstrates robustness to various failure modes, achieving full recovery of OOIs (where feasible) in every field run, and significantly outperforming our baseline.

Lane marker detection is a crucial challenge in developing self-driving cars. Despite significant research, large gaps remain between research and needs for fully autonomous driving. We highlight the limitations of present work and present a unified approach for robust and real-time lane marker detection. We present a multi-feature lane detection algorithm and give evidence why relying on one type of features can be harmful. We design a lane model using geometric constraints on lane shape and fit the lane model to the visual cues extracted. We improve the robustness of our algorithm by tracking lane markers temporally. We test our algorithm on KITTI dataset and show results that our algorithm can detect lane markers in presence of occlusions, sharp curves, and shadows.