Untethered soft actuators that operate without physical connections to external power sources or control system cables are an active area of research for their potential applications in human-robot interaction, haptic devices, and 4D printing due to their wide range of motion and high degrees of freedom. Among these actuators, liquid-to-gas phase change actuators have a simple structure and can generate higher output force using photothermal conversion. In this study, we propose a novel control method for untethered soft actuators by dyeing the low-boiling-point liquids used in liquid-to-gas phase change actuators. The dyed low-boiling-point liquids exhibit different light absorption spectra matching their color, enabling wavelength-selective actuation using light sources with corresponding wavelengths. We evaluated and discussed this control method in terms of the relationship between irradiance and output, actuation speed, and continuous operation. Additionally, we prototyped a robotic arm to demonstrate selective motion according to the light color.

This paper proposes a Winfree-model-type synchronization control method for a swimming modular robot. In addition to reconfigurability, fault tolerance, robustness, and adaptability are crucial characteristics for modular robots. Self-organization phenomena arising from physical interactions play an important role in achieving these capabilities. Building on the Winfree synchronization model, we develop a Winfree-model-type synchronization control approach and apply it to a swimming modular robot inspired by Volvocine algae. We experimentally verify that this method enables self-organized synchronization through physical interactions and contributes to stabilizing the posture of the robot colony. Furthermore, we demonstrate the fault tolerance of the system through experiments in which one module in the colony ceases operation. Extending the interaction-based philosophy demonstrated by TEGOTAE controllers to swimming modular robots, our proposed Winfree-type scheme functions independently of specific structural assumptions.

A novel path planning algorithm, RewardRRT, is proposed to address the challenge of Multi-degree-of-freedom robot path planning in narrow environments. In this approach, RewardRRT conceptualizes the sampling tree of Rapidly-exploring Random Trees (RRT) as an agent, assigning a reward value function to each sampled state. Simultaneously, the cumulative reward and reward increment are utilized as the state space, with a linear Kalman Filter applied to predict the state transitions. To enhance convergence speed, a bidirectional expansion strategy is implemented, wherein the tree with the lower cumulative reward prediction is prioritized for iteration. Here, the sampling bias is regulated using a sigmoid function and the prediction value. Finally, simulation tests in 4 distinct point cloud scenarios and a real experiment are conducted using a self-designed 21-degree-of-freedom wheeled humanoid robot. Compared to the best-performing algorithm from the Open Motion Planning Library (OMPL) in the same scenarios, RewardRRT achieves improvements in speed by 38.45%, 8.18%, 9.88%, and 14.98%, respectively. Furthermore, RewardRRT exhibits an average planning success rate of 88.25%, surpassing OMPL's best-performing algorithm by 29.75%. These results underscore the effectiveness of RewardRRT in solving path planning challenges in narrow environments.

Aerial transportation robots using suspended cables have emerged as versatile platforms for disaster response and rescue operations. To maximize the capabilities of these systems, robots need to aggressively fly through tightly constrained environments, such as dense forests and structurally unsafe buildings, while minimizing flight time and avoiding obstacles. Existing methods geometrically over-approximate the vehicle and obstacles, leading to conservative maneuvers and increased flight times. We eliminate these restrictions by proposing PolyFly, an optimal global planner which considers a non-conservative representation for aerial transportation by modeling each physical component of the environment, and the robot (quadrotor, cable and payload), as independent polytopes. We further increase the model accuracy by incorporating the attitude of the physical components by constructing orientation-aware polytopes. The resulting optimal control problem is efficiently solved by converting the polytope constraints into smooth differentiable constraints via duality theory. We compare our method against the existing state-of-the-art approach in eight maze- like environments and show that PolyFly produces faster trajectories in each scenario. We also experimentally validate our proposed approach on a real quadrotor with a suspended payload, demonstrating the practical reliability and accuracy of our method.

Autonomous navigation in unstructured, off-road environments remains a critical challenge due to complex terrain and limited prior knowledge. This letter presents a hierarchical framework to generate paths across locally flat regions and optimize spatio-temporal trajectories under vehicle kinodynamics. The first step employs a terrain-aware RRT* that integrates Bayesian-learning-guided sampling with a directional reachability index, enabling efficient exploration in complex environments while avoiding dead-end traps. The resulting path explicitly accounts for local roughness and slope. The second step applies a constrained iterative linear quadratic regulator (CiLQR) trajectory optimizer that refines the path within a dynamic flat-corridor, where terrain risk is embedded into the optimization. The optimizer enforces vehicle stability constraints and adaptively regulates speed based on terrain risk and curvature, yielding dynamically feasible trajectory. Extensive simulations in off-road scenarios demonstrate that the proposed framework outperforms state-of-the-art methods, producing feasible trajectories with up to 34 % lower vehicle jolt. Furthermore, real-world tests on a 40-ton articulated truck in the Alxa Desert validate the framework's robustness and practical effectiveness.

This paper addresses vision-based autonomous landing of quadrotor drones on moving platforms with uncertain motion. Vision is attractive due to its low weight, low cost, and ability to provide direct relative observations without global reference frames. However, traditional visual landing relies on requiring accurate estimation and tuning, which limits robustness. Deep reinforcement learning (DRL) offers a data-driven alternative but often degrades under motion uncertainty or intermittent visual loss from a limited field of view (FOV). The key challenge is active perception, maintaining visual observability of the landing pad under FOV constraints during aggressive maneuvers. To address this challenge, we propose a vision-based DRL framework that jointly learns perception, estimation, and control, guided by an active-perception reward that couples visibility maintenance with control performance for stable touchdown. Simulation results demonstrate robustness over visual servoing and existing DRL baseline, including landings on a platform moving at speeds up to 8 m/s under limited visibility. Real-world experiments further confirm the feasibility and stability of the proposed approach.