[CLSTR]
< Back to situations

Monitor this situation.

[SITUATION] · [QUIET] · [TECHNOLOGY]

3 clusters · 12 sources · 8 days · First seen · Last updated

Advancements in biomimetic and mobile robotics

Overview

Recent developments in biomimetic and soft robotics have introduced new capabilities for autonomous movement and specialized surveillance.

In China, a company developed the ‘BG-5 Golden Dragon Fish’, an underwater robot modeled after the golden arowana. Designed for discrete reconnaissance in shallow waters, the 63-centimeter device uses a multi-jointed tail to mimic natural fish movements, allowing for silent operation without traditional propellers. Advanced versions, such as the BG-5V, support coordinated multi-robot operations and complex route planning.

Concurrently, researchers at ETH Zurich’s Soft Robotics Lab have focused on mobile robotic hands. Engineers successfully trained a commercial WUJI robotic hand to crawl autonomously using deep reinforcement learning. This 818-gram unit, equipped with a Raspberry Pi, motion sensors, and a lithium-polymer battery, can navigate 14 different surfaces.

To overcome stability challenges caused by the hand’s asymmetrical, human-like design—which includes unequal finger lengths and an opposing thumb—the team utilized NVIDIA’s Isaac lab simulation platform. This training allows the device to lift individual digits to maintain balance and forward progress. Beyond locomotion, the system can perform fine motor tasks, such as pushing small objects or tapping keyboard keys, while balancing on its fingertips. The robot also features a recovery algorithm that allows it to right itself in 84% of hardware fall attempts. These advancements aim to improve robot utility in confined or difficult-to-access environments.

Entities

Soft Robotics Lab · ETH Zurich · Amirhossein Kazemipour · WUJI · Hehui Zheng

Timeline

  1. [TECHNOLOGY] 3 sources
    ETH Zurich researchers develop walking robotic hand

    ETH Zurich researchers have taught a standard, asymmetrical robotic hand to walk on its fingertips using machine learning and integrated sensors, allowing for autonomous movement across various terrains.

  2. [TECHNOLOGY] 7 sources
    ETH Zurich engineers develop autonomous mobile robotic hand

    Researchers at ETH Zurich have developed a mobile, self-contained robotic hand that uses reinforcement learning to crawl and balance across diverse surfaces using its own fingers.

  3. [TECHNOLOGY] 2 sources
    Robotics advancements feature biomimetic underwater drone and mobile robotic hand

    New robotic developments include a biomimetic Chinese underwater drone for silent surveillance and a mobile robotic hand from ETH Zurich designed for navigating tight spaces.

Sources

descopera.ro · hothardware.com · ithome.com · knowridge.com · nypost.com · offthepress.com · popsci.com · stiripesurse.ro · tech.wp.pl · tech360.tv · technologyreview.de · teknoblog.com

This summary has been updated 1 time: see revision history