1 April 2015 in Powder Technology
24 March 2022 in Ocean Engineering
24 June 2024 in IEEE Robotics and Automation Letters
Soft robotics, including soft grippers and manipulators, are increasingly required to function more adaptively and securely in unknown environments. However, conventional soft pneumatic grippers often grasp and hold objects via the active bending of multiple fingers, making it challenging to grip and hold, for example, spiny objects safely because of the potential damage to the pneumatic chambers. In this letter, inspired by the entanglement behavior of plant tendrils, we propose a soft, pneumatic, fail-safe gripper. The fail-safe property is guaranteed by two mechanisms, the passive entanglement and the redundant contacts. Specifically, the gripper exploits an array of tentacle fingers that actively extend and passively entangle for entanglement grasping. To understand the working principle, we studied the entanglement behavior of a single tentacle finger by both analytical modeling and finite element method simulation. We also characterized the griping capability of this gripper in a series of different grasping experiments. The gripper was demonstrated to be capable of grasping a range of different objects, from artificial ones with simple shapes to real-world targets with complex shapes. We finally showcased the fail-safe property of this gripper in a pick-and-lift task by comparing the performance of this gripper with that of a conventional gripper when the two grippers are manually damaged. This work paves the way for the development of soft pneumatic grippers with fail-safe and reliable grasping.
26 March 2024 in Biomimetic Intelligence and Robotics
28 April 2025 in IEEE Transactions on Intelligent Transportation Systems
Multiple autonomous underwater vehicles (MAUVs) cooperative systems are extensively extensively utilized for conducting search missions of wrecked aircraft or submarines across vast oceanic areas. Ensuring safe and efficient complete coverage path planning is a critical challenge for MAUVs in performing these search tasks. In this paper, we present an efficient MAUVs cooperative adaptive coverage path planning algorithm. Specifically, we introduce the neuronal activity reassigned (NAR) algorithm as the initial step to devise optimal navigation paths for a single AUV. Through the utilization of both the AUV steering number optimization strategy and the deadlock point escape algorithm based on the glasius bio-inspired neural network (GBNN), the path planning process is jointly optimized. Then, the adaptive balance evolution (ABE) algorithm is designed with the AUV energy constraints as a guide to match each AUV to a subtask sea area with task volume differentiation by dividing the task sea area in multiple iterations based on the residual capacity difference of individual AUVs. Finally, the simulation and experiment results demonstrate that the MAUVs cooperative adaptive coverage path planning algorithm proposed in this paper offers several advantages, including optimal task balance, extensive coverage, minimal path repetition rate, reduced occurrence of deadlock points, and decreased steering times. These benefits are instrumental in facilitating the MAUVs system’s efficient execution of refined search tasks.
20 July 2024 in Biomimetic Intelligence and Robotics