Matt Strong

Matt Strong

CS PhD at Stanford University

← Paper Digest

Design of a hybrid tendon-linkage finger with coupled and adaptive motion

Hyojae Kang, Giup Seo, Jae-Young Lee, Yeong Jae Park, Jongwoo Park, Hyunmin Do, So-Hee Park, Dongil Park, Jeongdo Ahn

November 2026

TL;DR A robot finger with four joints driven by three motors, where the fingertip joint follows the middle joint through a spring-loaded four-bar linkage and bends further once contact blocks the middle joint. The built finger tracks the predicted coupling with 2.54 degrees mean error and presses with 4.56 N at the fingertip.

What they do

Actuating every joint of a robot finger independently means more motors and more tendons routed through the proximal joints. This paper builds a finger with four mechanical degrees of freedom and three actuators, all housed on the palm side and none in the moving links. The fingertip (DIP) joint has no motor: it follows the middle (PIP) joint in free space and bends further on its own when contact stops the PIP joint.

How it works

  • Three transmissions. MCP abduction/adduction goes through a bevel gear. MCP and PIP flexion/extension each use a pair of opposing tendons pulled by a motor, gear train and ball screw, so both directions are actively driven without return springs.
  • Coupled DIP. A four-bar linkage ties DIP to PIP. Two of its links are joined by a torsion spring and held against a mechanical stopper, so they act as one rigid link during free motion.
  • Adaptive DIP. When contact constrains the PIP joint, further actuation rotates the two spring-connected links relative to each other, which adds DIP flexion.
  • MCP-PIP coupling. The PIP tendon passes through the MCP region, so MCP flexion also moves the PIP joint; the kinematic model accounts for this.

Results

  • Coupled PIP-DIP motion, measured with ArUco markers, matches the linkage prediction with 2.54 degrees mean absolute error and 3.60 degrees RMSE (71 measurements).
  • Maximum adaptive motion is within 3.63 degrees mean absolute error of the theoretical envelope (110 measurements), but falls short by up to 18.95 degrees, mostly beyond about 70 degrees of PIP flexion.
  • Mean peak fingertip force is 4.56 N over five trials (standard deviation 0.045 N), in one tested configuration. The finger weighs 193.6 g.
  • The finger takes different contact postures on four cylinders (7.9 to 90.2 mm diameter), two cuboids and one irregular object.

Limitations

Interference between linkage components prevents the full adaptive motion at some PIP angles. The MCP joint is not backdrivable. The force figure describes one configuration, not the maximum over the joint range. It is a single finger: object tests show contact postures, not grasps, and a thumb and multi-fingered hand are future work.

Figures

From Figure 1: Overall structure of the finger: (a) dimensions, (b) joints and phalanges, and (c) internal transmissi...
From Figure 1: Overall structure of the finger: (a) dimensions, (b) joints and phalanges, and (c) internal transmission components with detailed views.
From Figure 3: (a) Serial kinematic chain and coordinate frames, and (b) planar flexion transmission showing the MCP ...
From Figure 3: (a) Serial kinematic chain and coordinate frames, and (b) planar flexion transmission showing the MCP and PIP tendon displacements and the distal coupled-adaptive linkage.
From Figure 6: Measured coupled and adaptive PIP-DIP motion against the four-bar linkage prediction, plotted as cumul...
From Figure 6: Measured coupled and adaptive PIP-DIP motion against the four-bar linkage prediction, plotted as cumulative PIP and DIP flexion over PIP angle.
From Figure 8: Postures of the finger in contact with cylinders, cuboids and an irregularly shaped object.
From Figure 8: Postures of the finger in contact with cylinders, cuboids and an irregularly shaped object.
Abstract

Dexterous robotic fingers require coordinated joint motion and adaptability to object contact, while independently actuating multiple joints increases the complexity of the transmission mechanism. This paper presents a hybrid tendon--linkage robotic finger with four mechanical degrees of freedom driven by three actuators. The MCP abduction/adduction joint is driven through a bevel gear, while MCP and PIP flexion/extension are driven by separate bidirectional tendon transmissions. The DIP joint is not independently actuated but is coupled to the PIP joint through a four-bar linkage with a torsional spring and mechanical stopper. The tendon routing also introduces mechanical coupling between the MCP and PIP joints. Kinematic analysis was conducted to describe the coupled and adaptive motions, and experiments were performed to evaluate the joint motion characteristics. ArUco-based measurements showed that the coupled motion closely matched the theoretical prediction, whereas the adaptive motion was partially limited by interference between linkage components. The fabricated finger demonstrated different contact postures with cylindrical, cuboid, and irregularly shaped objects, and the mean peak fingertip force was 4.56 N. The proposed mechanism provides both coupled motion and contact-dependent adaptation without requiring an additional actuator for the DIP joint.

Summaries are written automatically from each paper's text. Figures and videos come from the paper and its project page; all credit goes to the authors.