DTC: Deep Tracking Control.

Legged locomotion is a complex control problem that requires both accuracy and robustness to cope with real-world challenges. Legged systems have traditionally been controlled using trajectory optimization with inverse dynamics. Such hierarchical model-based methods are appealing because of intuitive cost function tuning, accurate planning, generalization, and, most importantly, the insightful understanding gained from more than one decade of extensive research. However, model mismatch and violation of assumptions are common sources of faulty operation. Simulation-based reinforcement learning, on the other hand, results in locomotion policies with unprecedented robustness and recovery skills. Yet, all learning algorithms struggle with sparse rewards emerging from environments where valid footholds are rare, such as gaps or stepping stones. In this work, we propose a hybrid control architecture that combines the advantages of both worlds to simultaneously achieve greater robustness, foot-placement accuracy, and terrain generalization. Our approach uses a model-based planner to roll out a reference motion during training. A deep neural network policy is trained in simulation, aiming to track the optimized footholds. We evaluated the accuracy of our locomotion pipeline on sparse terrains, where pure data-driven methods are prone to fail. Furthermore, we demonstrate superior robustness in the presence of slippery or deformable ground when compared with model-based counterparts. Last, we show that our proposed tracking controller generalizes across different trajectory optimization methods not seen during training. In conclusion, our work unites the predictive capabilities and optimality guarantees of online planning with the inherent robustness attributed to offline learning.

Single-crystal ZrCo nanoparticle for advanced hydrogen and H-isotope storage.

Hydrogen-isotope storage materials are essential for the controlled nuclear fusion. However, the currently used smelting-ZrCo alloy suffers from rapid degradation of performance due to severe disproportionation. Here, we reveal a defect-derived disproportionation mechanism and report a nano-single-crystal strategy to solve ZrCo's problems. Single-crystal nano-ZrCo is synthesized by a wet-chemistry method and exhibits excellent comprehensive hydrogen-isotope storage performances, including ultrafast uptake/release kinetics, high anti-disproportionation ability, and stable cycling, far superior to conventional smelting-ZrCo. Especially, a further incorporation of Ti into nano-ZrCo can almost suppress the disproportionation reaction. Moreover, a mathematical relationship between dehydrogenation temperature and ZrCo particle size is established. Additionally, a microwave method capable of nondestructively detecting the hydrogen storage state of ZrCo is developed. The proposed disproportionation mechanism and anti-disproportionation strategy will be instructive for other materials with similar problems.

MXene-Reinforced Liquid Metal/Polymer Fibers via Interface Engineering for Wearable Multifunctional Textiles.

Stretchable conductive fibers are an important component of wearable electronic textiles, but often suffer from a decrease in conductivity upon stretching. The use of liquid metal (LM) droplets as conductive fillers in elastic fibers is a promising solution. However, there is an urgent need to develop effective strategies to achieve high adhesion of LM droplets to substrates and establish efficient electron transport paths between droplets. Here, we use large-sized MXene two-dimensional conductive materials to modify magnetic LM droplets and prepare MXene/magnetic LM/poly(styrene-butadiene-styrene) composite fibers (MLMS fibers). The MXene sheets decorated on the surface of magnetic LM droplets not only enhance the droplet adhesion to substrate but also bridge adjacent droplets to establish efficient conductive paths. MLMS fibers show several-fold improvements in tensile strength and elongation and a 30-fold increase in conductivity compared with pure LM-filled fibers. These conductive fibers can be easily woven into multifunctional textiles, which exhibit strong electromagnetic interference shielding and stable Joule heating performances even under large tensile deformation. In addition, other advantages of MLMS textiles, such as high gas/liquid permeability, strong chemical resistance (acid and alkaline conditions), high/low-temperature tolerance (-40-150 °C) and water washability, make them particularly suitable for wearable applications.