Novel Robotic Hand Features Reversible Fingers, Merging Mobility and Multi-Object Manipulation

Robotic Hand Combines Manipulation and Mobility with Reversible Fingers

Researchers have developed an innovative robotic hand featuring a novel reversible-finger design, which addresses long-standing limitations of traditional, human-like robotic hands. This new system integrates multiple functionalities into one platform.

This new system allows the hand to grasp from both sides, move autonomously, handle multiple objects, and dock itself, integrating various functions into one platform.

Traditional robotic hands often mimic the human hand's asymmetric, thumb-centric, and fixed design. While effective for many dexterous tasks, this approach presents limitations in symmetric grasps, multi-object handling, and navigation in confined spaces. The research team drew inspiration from nature's generalists, such as octopuses, to create a fully modular, symmetric hand. Its reversible fingers function both as grippers and as legs, enabling locomotion. Additionally, the palm can detach and reattach autonomously using a magnetic docking mechanism.

Design and Control Mechanisms

To unify manipulation and mobility, the researchers created a co-optimized system involving hardware, motion planning, and control.

A virtual "grasp taxonomy" was developed to facilitate grasping during locomotion. Instead of assigning entire fingers to specific tasks, finger segments were dynamically mapped to functional roles, allowing simultaneous crawling and grasping.

A dynamical systems approach generates real-time velocity fields for obstacle-aware crawling and docking. Finger trajectories are guided by a central pattern generator (CPG), which produces stable, cyclic joint motions adaptable to desired directions and terrain.

The hand's structure was optimized via a genetic algorithm, evaluating various finger placements, roles, and locomotion gaits. Simulations indicated that configurations with three to six fingers offered optimal performance, balancing dexterity and mobility while minimizing weight and collision risks. The prototype uses modular, reversible fingers actuated by commercial servos, mounted on a detachable palm with neodymium magnets and a motorized bolt. A vision system supports real-time object detection, with position-based control ensuring precise tracking.

Key Capabilities and Results

The robotic hand demonstrated several core functionalities through validation experiments:

• Optimized Grasp-and-Crawl: The co-designed system confirmed that three to six fingers provide the best balance, with more than five fingers yielding diminishing returns due to increased mass and collision potential.

• Broad Grasping: The symmetric fingers performed all 33 grasp types in the Feix GRASP taxonomy, including pinch, tripod, and spherical grasps. The hand could handle up to four objects concurrently and lift loads up to 2 kg. Its joint motion range exceeded that of a human hand, expanding its workspace.

• Autonomous Role Switching: The hand successfully detached from its arm, crawled to retrieve scattered items (including stacking them), and re-docked without external intervention, demonstrating independent operation and reintegration capabilities.

• Resilience and Self-Recovery: Its reversible design enabled grasping from either side and allowed the hand to self-right using its fingers if inverted, showcasing adaptability in unstructured environments.

Broader Implications

This research deviates from the traditional approach of mimicking biological systems, instead exploring new design spaces by integrating functions typically separate. Performance improvements include a 5–10% increase in crawling efficiency over asymmetric designs, the ability for any finger pair to act as opposable thumbs, simpler multi-object planning, and reduced system complexity through shared hardware.

Potential applications span disaster response, industrial inspection, assistive robotics, and prosthetics, where autonomous manipulation and movement in complex environments are critical. The design also offers possibilities for prosthetic technologies or supernumerary limbs requiring adaptability and intuitive control.

The research challenges the notion that robotic hands must imitate human anatomy, demonstrating that prioritizing function, symmetry, reversibility, and modularity can lead to a more versatile and self-sufficient robotic platform.

This integrated system blurs the lines between manipulator and mobile robot, potentially indicating a future direction for robotics.