Safe navigation in dynamic environments is challenging when system dynamics are unknown and actuator inputs are limited. Existing methods either rely on accurate models, require online optimization, or do not explicitly account for input constraints. This paper presents a real-time control framework for unknown Euler-Lagrange systems that guarantees finite-time reach-avoid-stay (FT-RAS) specifications while respecting actuator limits. We extend the spatiotemporal tube (STT) framework by incorporating input constraints into the controller design and derive offline-verifiable feasibility conditions that relate the available control authority to the tube design and uncertainty bounds. The resulting framework is approximation-free and computationally efficient, making it suitable for real-time implementation. The proposed approach is validated through simulations on a mobile robot, a quadrotor, and a spacecraft, together with hardware experiments on a mobile robot, demonstrating safe navigation while satisfying actuator constraints.
This letter proposes the Predictive Lightweight Multi-Agent Reinforcement Learning (PL-MARL) framework to ensure resilient coverage in bandwidth-constrained UAV swarms. To counter coordination collapse caused by sparse signaling and information aging, we introduce a Kinematic-Awa…
Action-conditioned video world models predict future observations from an initial observation and an action signal. In robotics, actions influence future observations through two distinct processes: they are first realized into robot motion by the robot body and controller, and t…
Accurate yet low-latency depth is essential for radar-camera perception in autonomous systems. Cameras provide rich appearance but lack metric scale, whereas automotive radar offers metric range but is sparse and noisy. Many pipelines are multi-stage or depend on auxiliary annota…
Contact-rich robot manipulation requires physical interaction cues that are often invisible to cameras, making tactile sensing essential for robust control. However, scaling visuo-tactile robot learning remains difficult because real tactile interaction data are expensive to coll…
In production processes for consumer products, assembly instructions are essential not only for planning but also for executing the production process. Likewise in robotics, it is crucial for an assembly robot to understand how components fit together and can be assembled. To fac…
Geometry Foundation Models (GFMs) have substantially advanced monocular 3D reconstruction, yet extending this capability to 4D dynamic understanding remains a fundamental challenge. Most existing motion perception methods (e.g., sparse tracking, dense point-wise flow) treat motio…