Source-linked AI summary
Design of a Fully Actuated 4-DOF Robotic Finger With Joint-Specific Hybrid Remote Actuation
Hyojae Kang, Hyun-mok Jung, Joonho Lee, Dongil Park, Hyunmin Do, Jongwoo Park, Jeongdo Ahn
TL;DR
Fully actuated multi-DoF fingers face actuator-packaging and distal-wire routing challenges. This paper presents a four-DoF finger with rigid-link MCP actuation and closed-loop RCJ wire transmissions at PIP and DIP. Experiments showed the intended distal-transmission decoupling with the MCP fixed and measured mean peak fingertip forces of 21.28 N, 9.22 N, and 5.75 N under isolated MCP, PIP, and DIP actuation.
Problem
Fully actuated multi-DoF fingers require independent joint inputs, while added actuators and serial distal wire routing raise packaging and unintended-coupling challenges.
Source: §1 · p. 1Method
The paper combines two rigid-link transmission sets at the 2-DOF MCP joint with independent closed-loop RCJ wire transmissions at the PIP and DIP joints.
Source: §1 · pp. 1–2Results
21.28 N, 9.22 N, and 5.75 N were the mean peak fingertip forces under isolated MCP, PIP, and DIP actuation, respectively; distal transmissions showed no appreciable cross-displacement under isolated actuation with the MCP fixed.
Source: §5.2 · p. 8, §6 · p. 9Takeaways & Limitations
The experiments indicate that this architecture can preserve independent joint-level actuation while relocating distal actuators away from the finger links.
Source: §6 · p. 9Future work will reduce wire-system friction and transmission losses, quantify efficiency, hysteresis, and backdrivability, and extend the module to a five-finger hand for systematic grasping and manipulation evaluation.
Source: §6 · p. 9
Abstract
from arXiv · showhide
This paper presents a fully actuated 4-DOF robotic finger using a joint-specific hybrid remote-actuation architecture. The metacarpophalangeal (MCP) joint is driven by two coordinated rigid-link transmission sets, whereas the proximal interphalangeal (PIP) and distal interphalangeal (DIP) joints are independently actuated by closed-loop wire transmissions incorporating circular rolling-contact joints (RCJs). A larger transmission radius is used at the PIP joint than at the DIP joint. The RCJ wire geometry maintains the total wire-loop length during joint rotation, and the DIP wire routing is designed so that PIP motion does not affect its differential actuation for a fixed MCP configuration. The distal wire transmission, MCP linkage, and fingertip kinematics are analytically modeled. In experiments, the transmission behavior was quantitatively evaluated from the ball-screw displacements measured using ArUco-marker tracking. MCP actuation produced measurable displacements of the PIP and DIP transmission units, whereas isolated PIP and DIP actuation with the MCP fixed supported the intended mechanical decoupling between the distal transmissions. The mean peak fingertip forces under isolated MCP, PIP, and DIP actuation were 21.28 N, 9.22 N, and 5.75 N, respectively. The resulting finger postures were also examined using objects of different geometries and sizes.
1 Introduction
The paper addresses the packaging and routing challenges of fully actuated multi-DoF fingers with a joint-specific hybrid remote-actuation architecture. It combines rigid links at the 2-DOF MCP joint with closed-loop wires at the PIP and DIP joints, then evaluates transmission behavior and fingertip forces.
1 Introduction
- Fully actuated multi-DoF fingers require independent inputs, but increasing actuated joints also increases actuator and transmission components.Packaging is restrictive toward distal phalanges, and distal actuator placement adds carried mass and structural volume.Source: §1 · p. 1
- Serial distal wires must support bidirectional actuation and avoid unintended coupling when they traverse moving proximal joints.The proposed routing is designed to decouple PIP and DIP transmissions for a fixed MCP configuration.Source: §1 · pp. 1–2
- The finger architecture retains independent actuation of all four DoFs without actuators in distal finger links, while distal wire routing prevents differential coupling between PIP and DIP transmissions for a fixed MCP configuration.Experiments quantified MCP-induced displacement of distal transmission units and evaluated mechanical decoupling for a fixed MCP configuration, along with fingertip force and object-interaction behavior.Source: §1 · p. 2
- The design combines two coordinated rigid-link sets at the 2-DOF MCP joint with remote closed-loop wire transmissions at the serial PIP and DIP joints.The four-DoF architecture retains independent joint actuation while keeping actuators out of the distal finger links.Source: §1 · p. 2
2 Related Works
Related work contrasts underactuated and fully actuated fingers, then reviews remote and hybrid transmissions and rolling-contact joints. The proposed design assigns rigid links proximally and closed-loop wires to the distal joints.
2 Related Works
- Underactuated fingers coordinate multiple joint motions with fewer actuators, whereas fully actuated systems assign independent actuation to each controlled DoF.The additional actuators and transmission components make packaging a major design consideration in multi-DoF fingers.Source: §2 · p. 2
- Wire-driven fingers place actuators away from distal joints, but tension management and routing geometry affect their performance.For serial joints, changing proximal wire paths can cause unwanted coupling; routing through revolute-joint centers has been used to address this issue.Source: §2 · p. 2
- Rolling-contact joints let designers shape joint trajectory and transmission geometry together, and have been used in several finger and manipulator mechanisms.The proposed finger uses circular RCJs in closed-loop wires at PIP and DIP, with the DIP path routed through the PIP structure to decouple distal transmissions for a fixed MCP configuration.Source: §2 · p. 2
- The proposed architecture assigns rigid-link transmission to the proximal 2-DOF MCP joint and closed-loop wire transmission to the serial PIP and DIP joints.This avoids additional tendon-routing elements around the MCP while keeping distal actuators outside the phalanges.Source: §2 · p. 2
3 Mechanical Design
The mechanical design provides four independently actuated DoFs using rigid-link MCP transmission and remote wire transmissions for PIP and DIP. The distal routing uses RCJs and separate wire paths to maintain bidirectional actuation and distal-joint decoupling when the MCP is fixed.
3 Mechanical Design
- Four motors actuate MCP abduction/adduction and flexion/extension, plus PIP and DIP flexion/extension.The module is less than 200 mm long; the finger portion is 19.00 mm wide and 16.09 mm thick.Source: §3 · p. 2
- Two rigid-link transmission sets use coordinated ball-screw motion to generate MCP abduction/adduction and flexion/extension.Each set connects a ball-screw-side universal joint to a spherical joint linked to the MCP structure.Source: §3 · pp. 2–3
- Closed-loop wire transmissions and circular RCJs drive the serial PIP and DIP joints, with a larger effective transmission radius at PIP than DIP.The larger PIP radius provides a larger wire moment arm at that joint.Source: §3 · p. 3
- Three wire systems serve the distal mechanisms: RCJ constraint wires, a PIP actuation wire, and a DIP actuation wire.The actuation wires are 0.45 mm-diameter 7×19 stranded SUS304 stainless-steel cables.Source: §3 · p. 3
- Separate distal differential paths route the DIP wire through the PIP structure, so PIP motion does not produce differential DIP actuation with the MCP fixed.MCP motion changes the proximal routing lengths of both distal actuation wires because they pass through the moving MCP structure.Source: §3 · p. 4
4 Transmission and Kinematic Analysis
The analysis models the RCJ wire paths and MCP rigid-link transmissions, including how their geometry maps joint motion to actuator displacement. It also formulates fingertip kinematics and evaluates the reachable workspace over the selected joint ranges.
4.1 RCJ Wire Transmission
- For fixed MCP configuration, the ideal RCJ routing yields a diagonal PIP–DIP transmission relation, so motion at either joint does not generate differential actuation at the other.The opposing wire branches compensate to keep total loop length unchanged during joint rotation; the DIP passage through the PIP structure remains constant during PIP rotation.Source: §4.1 · pp. 4–5
4.2 MCP Rigid-Link Transmission
- The MCP rigid-link model determines ball-screw displacements d1 and d2 from the MCP angles q1 and q2 using the positions of the linkage joints and constant-length constraints.The model first obtains P3 and P4 from the MCP geometry, then solves for the corresponding actuator displacements.Source: §4.2 · p. 5
4.3 Finger Kinematics
- The forward-kinematic formulation uses Craig’s modified Denavit–Hartenberg parameters, with RCJ coordinates q3 and q4 corresponding to twice those relative joint rotations.The reachable fingertip workspace is calculated over the joint space using the selected operating ranges.Source: §4.3 · pp. 5–6, Figure 6 · p. 6
5 Experiments
Experiments evaluated transmission behavior, fingertip force, and object-interaction motion. ArUco-tracked ball-screw displacements were used to assess MCP-induced distal-path changes and PIP–DIP decoupling with the MCP fixed.
5.1 Motion and Transmission Evaluation
- Transmission tests tracked ball-screw displacement with ArUco markers while powering only the motors needed for the target motion and leaving other motors backdrivable.The tests examined MCP-induced changes in distal wire paths and transmission decoupling during isolated PIP or DIP motion.Source: §5.1 · p. 6, §5 · p. 6
- At maximum MCP ball-screw displacement, DIP and PIP units moved 2.931 mm and 1.197 mm; isolated distal-joint tests showed no visible motion, supporting decoupling.The CAD model predicted a 3.84 mm routing-length change; wire tension and elastic deformation were cited as nonideal factors affecting observed motion.Source: Figure 7 · p. 7, §5.1 · pp. 7–8
5.2 Fingertip Force Measurement
- Mean peak forces were 21.28 N (MCP), 9.22 N (PIP), and 5.75 N (DIP) during isolated actuation; MCP force depends on finger configuration.Figure 9 presents the measured force profiles from three trials and their mean.Source: Figure 9 · p. 8, §5.2 · p. 8
5.3 Object Interaction
- Postures were examined with five cylindrical objects, two cuboids, and two irregular objects, adjusting joint configurations to match object diameter or geometry.Figure 10 shows the resulting object-interaction postures.Source: Figure 10 · p. 8, §5.3 · p. 8
6 Conclusion
The paper presents and experimentally validates a joint-specific hybrid remote-actuation architecture for a fully actuated 4-DOF robotic finger. Analytical results and experiments support its wire-loop-length preservation and distal-transmission decoupling, while future work targets transmission characterization and whole-hand evaluation.
6 Conclusion
- The RCJ-based wire transmission preserves total agonist–antagonist wire-loop length during joint rotation and intrinsically decouples PIP and DIP transmissions for a fixed MCP configuration.The architecture uses rigid-link transmission at the 2-DOF MCP joint and closed-loop wire transmission at the serial PIP and DIP joints.Source: §6 · p. 8
- Experiments confirmed that MCP actuation displaced the PIP and DIP transmission units, while isolated distal actuation with the MCP fixed produced no appreciable cross-displacement.Displacements were measured using ArUco-based tracking.Source: §6 · pp. 8–9
- 21.28 N, 9.22 N, and 5.75 N were the mean peak fingertip forces under isolated MCP, PIP, and DIP actuation, respectively; contact postures varied across object geometries.The experiments also found no appreciable cross-displacement between distal transmissions during isolated PIP or DIP actuation with the MCP fixed.Source: §6 · p. 9
- Future work will reduce wire-transmission friction and losses, quantify efficiency, hysteresis, and backdrivability, and extend evaluation to a complete five-finger hand.Planned evaluation includes systematic grasping experiments and assessment of object-grasping and tool-manipulation capabilities.Source: §6 · p. 9