Zhao Luo | Searcharxiv

arXiv2026

Mechanics of hierarchical twisted and coiled polymer artificial muscles: Decoupling force from kinematic limits

Thermally actuated twisted and coiled polymer (TCP) artificial muscles exhibit exceptional specific work capacities but are limited by an inherent competition between load-bearing capacity and actuation stroke. To address this limitation, we investigate a hierarchical helical structure designed to decouple force generation from kinematic limits. We propose a coupled thermo-mechanical model incorporating inter-filamentary contact mechanics and geometric nonlinearities to predict the assembly's equilibrium response. The results indicate that this hierarchical topology significantly amplifies isometric actuation stress compared to monofilament baselines, while maintaining a biological-like contraction stroke of approximately 22%. A critical topological threshold governed by the balance between cooperative load-sharing and geometric confinement is identified. Beyond an optimal bundle complexity, the geometric jamming dominates, as excessive inter-filamentary friction hinders actuation. Furthermore, we elucidate a stiffness-stroke synergy in homochiral configurations, where high helical angles amplify the thermal untwisting torque to overcome increased structural rigidity. Crucially, the volumetric energy density exhibits scale invariance regarding the hierarchical radius, implying that absolute force output can be linearly scaled through geometric upsizing without compromising efficiency. These findings provide a mechanics-based rationale for the structural programming, demonstrating that soft actuator performance limits are dictated by topological order rather than intrinsic material properties.

cond-mat.soft↗

arXiv2024

Distilling Channels for Efficient Deep Tracking

Deep trackers have proven success in visual tracking. Typically, these trackers employ optimally pre-trained deep networks to represent all diverse objects with multi-channel features from some fixed layers. The deep networks employed are usually trained to extract rich knowledge from massive data used in object classification and so they are capable to represent generic objects very well. However, these networks are too complex to represent a specific moving object, leading to poor generalization as well as high computational and memory costs. This paper presents a novel and general framework termed channel distillation to facilitate deep trackers. To validate the effectiveness of channel distillation, we take discriminative correlation filter (DCF) and ECO for example. We demonstrate that an integrated formulation can turn feature compression, response map generation, and model update into a unified energy minimization problem to adaptively select informative feature channels that improve the efficacy of tracking moving objects on the fly. Channel distillation can accurately extract good channels, alleviating the influence of noisy channels and generally reducing the number of channels, as well as adaptively generalizing to different channels and networks. The resulting deep tracker is accurate, fast, and has low memory requirements. Extensive experimental evaluations on popular benchmarks clearly demonstrate the effectiveness and generalizability of our framework.

cs.CV↗