ORBIT: A Modular Multi-finger Robotic Hand Platform with Rotational Finger Reconfiguration

Benito Ribadeneira; Siqi Shang; Lillian Chin

cs.ROOct 23, 2026

ORBIT: A Modular Multi-finger Robotic Hand Platform with Rotational Finger Reconfiguration

Authors: Benito Ribadeneira, Siqi Shang, Lillian Chin

Organizations: The University of Texas at Austin, TX, USA

Abstract

Despite the popularity of multi-fingered robotic hands for dexterous manipulation, there is no systematic platform to compare different hand morphologies against one another. We present ORBIT (Orbital Reconfigurable Base-actuated Integrated Theta-motion Hand), a modular platform designed to enable principled exploration of multi-fingered hand morphology. The core contribution of ORBIT is a rotating knuckle mechanism, where independent fingers travel around a cylindrical palm via a belt drive. This enables ORBIT to perform quick recon- figuration (< 1 s) of finger layout without hand disassembly. ORBITs a robust system, with finger average angular velocity of 1.57 rad / s1 and a root finger force resistance of 50 N. Using a three fingers instantiation of ORBIT, we demonstrate an initial pilot grasping study, comparing anthropomorphic, isometric, and Barrett-style finger layouts across eleven objects of varied size, shape, and texture. We further demonstrate how ORBITs nonanthropomorphic orbital DOF enables the achievement of tasks that would be difficult to perform with fixed-knuckle designs, such as cube face reorientation, lid threading, and precision pinching of sub-centimeter objects. These results establish ORBIT as a viable platform for direct apples to apples comparison of multi-fingered hand morphologies for specific manipulation tasks

Figures & tables

Fig. 1 : Overview of orbit mounted on a robot arm. The same orbit hardware allows us to use rotational movement to switch between different finger morphologies, including: (A) Anthropomorphic style, where two fingers oppose an offset thumb, (B) Isometric, where fingers are evenly distributed around the palm, (C) and Barrett style, where two fingers oppose a central thumb. Scale bars are $50\text{\,}\mathrm{mm}$ .

Fig. 2 : Conceptual overview of orbit and its anthropomorphic hand morphology instantiation. Each of orbit ’s DOFs have a biological analogue: (1) root finger displacmeent, (2) abduction / adduction, and (3) flexion / extension. Similarly, the finger joints correspond to the Distal interPhalangeal (DIP), Proximal InterPhalangeal (PIP), MetaCarpoPhalangeal (MCP) joints of the finger, while the soft palm mimics the thenar, hipothenar and Interphalangeal pads on the human palm. Scale bars are $50\text{\,}\mathrm{mm}$

Fig. 3 : Demonstration of orbit ’s modular architecture. From left to right, we demonstrate how we can swap a standard finger design with a shorter thumb finger design without needing to disassemble the entire hand. We also demonstrate conceptual pictures of how additional finger modules can be added or removed from the rotating knuckle mechanism, creating new hand layouts with the same base architecture. Scale bars are $50\text{\,}\mathrm{mm}$

Fig. 4 : Overview of the specific orbit instantiation we use in the paper. From left to right, general overview of orbit as a whole, exploded view of the two main subassemblies, and detailed information of the rotating knuckle mechanism and finger design. Scale bars are $50\text{\,}\mathrm{mm}$

Fig. 5 : Kinematic diagram of the finger design with the two cascaded four-bar linkage. Red links are driven inputs, blue links are grounded, green links are couplers and $\theta$ is the servo-input angle. Scale bars are $50\text{\,}\mathrm{mm}$

DoF	Metric	Result
DoF 1	Position accuracy	Within $\pm 2.0^{\circ}$
DoF 1	Force resistance	$49.46\pm 0.9\,$ \mathrm{N}$$
DoF 1	Angular velocity	$1.57\,$ \mathrm{rad}$$
DoF 2	Position accuracy	$<2^{\circ}$
DoF 2	Travel time	$0.53\pm 0.01\,$ \mathrm{s}$$
DoF 3	Position accuracy	$<2^{\circ}$

TABLE I : Consolidated orbit performance characterization.

Fig. 6 : Position accuracy characterization of the three DOF’s of orbit (A) DOF 1 DC motor feedback vs video tracking vs target angle. (B) DOF 2 servo motor feedback vs target angle around the video tracked path. (C) DOF 3 Video tracking vs Theoretical model path.

Fig. 7 : Grip force experimental setup for the three hand configurations with the two load cell presentations: (A) a spherical layout and (B) a flat layout. Scale bars are $50\text{\,}\mathrm{mm}$ .

Load case	Anthropomorphic	Isometric	Barrett
Spherical	$2.02\pm 0.18$	$1.64\pm 0.15$	$1.47\pm 0.37$
Flat	$2.42\pm 0.16$	$1.66\pm 0.16$	$2.20\pm 0.21$

TABLE II : Grip force (N) for each hand configuration and load case after a

10\text{\,}\mathrm{s}

settling window. Mean

\pm

standard deviation over three trials

Fig. 8 : (A) Manipulation target overview. Objects 1-11 were used to test grasp success. In numerical order, the objects are: Pringles can, hat, cracker box, M3x $5\text{\,}\mathrm{mm}$ screw, volleyball, five $2\text{\,}\mathrm{cm}$ cubes, earbud, guiro, pan, glass, drill. Objects 12 ( $5\text{\,}\mathrm{cm}$ cube) and 13 (peanut butter jar) were used to demonstrate the non-anthropomorphic ability of finger displacement. (B) Demonstration of five cubes lifted under isometric layout. (C) Guiro pick-up with the three configurations, showing a partially accepted, a failed and a successful pick-up, respectively. Scale bars are $50\text{\,}\mathrm{mm}$

Object	Challenge	A	I	B
1	Tall, light	✓	✓	✓
2	Large brim	✓	✓	✓
7	Small, smooth	✓	✓	✓
10	Delicate	✓	✓	✓
11	Heavy	✓	✓	✓
6	Unstable	Four	✓	✓

TABLE III: Grasping success for each hand configuration. Objects are described in Fig. 8 . A = Anthropomorphic, I = Isometric, B = Barrett. ✓ = success, ✗ = failure.

Fig. 9 : Special tasks using the dynamic orbital DOF 1 (A) Sub-centimeter object pick up (M3x5mm screw), (B) Fast cube orientation, (C) Jar-lid unscrewing.

Robotic Hand Design

Robotic hand design develops mechanical structures, actuation, and sensing for grasping and dexterous manipulation across real-world tasks.

See all 70 papers

Modular Anthropomorphic Hand Design via Multi-Parameter Finger Benchmarking and Selection
Jun 10, 2026Yu Zhang, Huijiang Wang, Josie HughesRobotic Hand Design Robotic Grasping
The CREATE Lab, Institute of Mechanical Engineering, Swiss Federal Institute of Technology in Lausanne (EPFL), CH-1015, Lausanne, Switzerland.
A Reconfigurable Dual-Opposition Architecture for Single-Hand Assembly and Manipulation
Sep 19, 2026William Su, Yunosuke Nakamura, Yixiao Wang +6Robotic Hand Design In-Hand Manipulation
Aerospace Engineering Program, University of California, Berkeley, Berkeley, CA 94720, USA. · Department of Mechanical Engineering, University of California, Berkeley, Berkeley, CA 94720, USA. · Department of Mechanical Engineering, National University of Singapore, Singapore 117575.
Handroid: Bridging Dexterous Hand and Humanoid
Jul 17, 2026Ruogu Li, Chenyang Ma, Sikai Li +6Humanoid Loco-Manipulation Robotic Hand Design
University of North Carolina at Chapel Hill · Stanford University

Jun 10, 2026J/K move · Enter open · S save

Modular Anthropomorphic Hand Design via Multi-Parameter Finger Benchmarking and Selection

Yu Zhang, Huijiang Wang, Josie Hughes

The CREATE Lab, Institute of Mechanical Engineering, Swiss Federal Institute of Technology in Lausanne (EPFL), CH-1015, Lausanne, Switzerland.

Robotic Hand Design Robotic Grasping Dexterous Manipulation Robotics Robotic Manipulation

Open paper AskPDF arXiv

Designing anthropomorphic dexterous robotic hands remains challenging as the design space straddles morphology, actuation, and sensing properties, and performance metrics span both task-dependent and task-agnostic. Existing optimization methods are often unstructured or consider only a single performance metric, limiting systematic comparison and targeted refinement. While the design considerations of the entire hand are significant, the individual finger properties play a key role in dexterity. By developing a robotic hand platform where fingers can be modularly integrated into a full teleoperated hand, we propose that optimizing the fingers can significantly improve overall hand performance. This approach enables rapid screening of different finger-level prototypes through a number of quantitative benchmarks before their integration into the hand for task-level validation. Candidate finger designs (incorporating variations in joint, bone, skin, and sensor placement) are assessed using both mechanism-oriented and task-relevant metrics, which establish a quantitative link between component design and full hand embodiment. The framework is validated through the development of an anthropomorphic robotic hand with optimized fingers, demonstrating how these fingers enable performance improvements across tasks, including multi-object grasping and light bulb screwing.

Abstract

Robotic Hand Design

Explore similar work

Modular Anthropomorphic Hand Design via Multi-Parameter Finger Benchmarking and Selection

A Reconfigurable Dual-Opposition Architecture for Single-Hand Assembly and Manipulation

Handroid: Bridging Dexterous Hand and Humanoid