DynaNav: Dynamic Feature and Layer Selection for Efficient Visual Navigation
This addresses efficiency and interpretability issues for robotics and embodied AI in resource-tight scenarios, representing a strong specific gain.
The paper tackled the problem of high computational overhead and lack of interpretability in visual navigation models by proposing DynaNav, which dynamically selects features and layers based on scene complexity, resulting in a 2.26x reduction in FLOPs, 42.3% lower inference time, and 32.8% lower memory usage while improving navigation performance.
Visual navigation is essential for robotics and embodied AI. However, existing foundation models, particularly those with transformer decoders, suffer from high computational overhead and lack interpretability, limiting their deployment in resource-tight scenarios. To address this, we propose DynaNav, a Dynamic Visual Navigation framework that adapts feature and layer selection based on scene complexity. It employs a trainable hard feature selector for sparse operations, enhancing efficiency and interpretability. Additionally, we integrate feature selection into an early-exit mechanism, with Bayesian Optimization determining optimal exit thresholds to reduce computational cost. Extensive experiments in real-world-based datasets and simulated environments demonstrate the effectiveness of DynaNav. Compared to ViNT, DynaNav achieves a 2.26x reduction in FLOPs, 42.3% lower inference time, and 32.8% lower memory usage, while improving navigation performance across four public datasets.