Search

Robot task planning is an important problem for autonomous robots in long-horizon challenging tasks. As large pre-trained models have demonstrated superior planning ability, recent research investigates utilizing large models to achieve autonomous planning for robots in diverse tasks. However, since the large models are pre-trained with Internet data and lack the knowledge of real task scenes, large models as planners may make unsafe decisions that hurt the robots and the surrounding environments. To solve this challenge, we propose a novel Safe Planner framework, which empowers safety awareness in large pre-trained models to accomplish safe and executable planning. In this framework, we develop a safety prediction module to guide the high-level large model planner, and this safety module trained in a simulator can be effectively transferred to real-world tasks. The proposed Safe Planner framework is evaluated on both simulated environments and real robots. The experiment results demonstrate that Safe Planner not only achieves state-of-the-art task success rates, but also substantially improves safety during task execution. The experiment videos are shown in https://sites.google.com/view/safeplanner ., Comment: 9 pages, 6 figures

Imitation learning (IL) has achieved considerable success in solving complex sequential decision-making problems. However, current IL methods mainly assume that the environment for learning policies is the same as the environment for collecting expert datasets. Therefore, these methods may fail to work when there are slight differences between the learning and expert environments, especially for challenging problems with high-dimensional image observations. However, in real-world scenarios, it is rare to have the chance to collect expert trajectories precisely in the target learning environment. To address this challenge, we propose a novel robust imitation learning approach, where we develop an inverse dynamics state representation learning objective to align the expert environment and the learning environment. With the abstract state representation, we design an effective reward function, which thoroughly measures the similarity between behavior data and expert data not only element-wise, but also from the trajectory level. We conduct extensive experiments to evaluate the proposed approach under various visual perturbations and in diverse visual control tasks. Our approach can achieve a near-expert performance in most environments, and significantly outperforms the state-of-the-art visual IL methods and robust IL methods.

Sensitivity to molecular chirality is crucial for many fields, from biology and chemistry to the pharmaceutical industry. By generating superchiral light, nanophotonics has brought innovative solutions to reduce the detection volume and increase sensitivity at the cost of a non-selectivity of light chirality or a strong contribution to the background. Here, we theoretically propose an achiral plasmonic resonator, based on a rectangular nanoslit in a thin gold layer behaving as a magnetic dipole, to generate a tunable nanosource of purely superchiral light. This nanosource is free of any background, and the sign of its chirality is externally tunable in wavelength and polarization. These properties result from the coupling between the incident wave and the magnetic dipolar character of our nano-antenna. Thus, our results propose a platform with deep subwavelength detection volumes for chiral molecules in particular, in the visible, and a roadmap for optimizing the signal-to-noise ratios in circular dichroism measurements to reach single-molecule sensitivity.