Source-linked AI summary
Development of a Voice-Controlled Tendon-Driven Bionic Hand
Urja Kohli, Shagata Chanda, Kritika Gandhi, Charu Nigam, Aditi Surya Kamal, Pooja Bhati
TL;DR
The paper addresses the effect of upper-limb amputation on daily tasks by developing an automated tendon-driven bionic hand. Its open-loop, voice-controlled design produces synchronized and selective finger movement with repeatable operation, although mechanical constraints and tendon-routing variation affect some motion details.
Problem
Upper-limb amputation greatly affects daily activities such as holding and manipulating objects, motivating assistive technologies.
Method
The system uses predefined input commands in an open-loop control system and tendon-driven actuation with passive forces returning the fingers to their original position.
Results
The hand achieved synchronized and selective movement, while tendons moved through actuation cycles without visible slack, breakage, or loss of tension; voice-based control was successfully used as an input device.
Takeaways & Limitations
The developed bionic hand provides repeatable, constant, smooth, and stable motion patterns for prosthetic applications.
Takeaways & Limitations
Variation in tendon routing and mechanical tolerances affects some motion curves, although these constraints do not impact general operation.
Abstract
from arXiv · showhide
The impairment of the hands can seriously affect the abilities of every individual to perform the every-day activity, so the design of stable and controllable support devices is a significant field of study. This paper is about the design and implementation of an automated bionic hand which is dedicated to the coordinated finger movement through the simplified and efficient actuation mechanism. The method that the proposed system was designed on is the tendon-based method whereby the servo motors generate the movement of the fingers, with assistance of the angular control which is calibrated. An actuation is controlled by a microcontroller that will be programmed by use of an Arduino-based microcontroller to carry out programmed gestures that include open hand, fist, pinch and half flexion. It has an interface that is voice command enabled to make it easy to interact with a Bluetooth based sender receiver architecture which offers an option of executing trained commands which are immediately converted to finger actions. To explore the motions behavior, finger coordination and control response to the input, the behavior of the experiment system is tested. The actuation of the fingers was found to take a total of about 7-8 seconds to achieve full flexion of all fingers in a sequence. The system showed repetitive and constant motion throughout several actuation cycles without loss of any apparent tension or precision of control. There was a stable grasp of objects of different shapes and sizes, which implied consistent coordination between the fingers. These findings indicate that the proposed system offers predictable and steady control behavior and has a simple and efficient mechanical and control architecture.
1 Department of Mechanical and Automation Engineering, Indira Gandhi Delhi Technical University for Women, Delhi, India
The paper addresses the need for functional, affordable prosthetic hands by proposing a lightweight voice-controlled bionic hand that combines anthropometric modelling, 3D-printed PLA construction, and tendon-driven servo actuation.
- Upper-limb amputation can substantially impair a person’s ability to perform daily tasks and manipulate objects.
- Existing prosthetic designs may not fit users outside the Western populations whose anthropometric data informed many commercial systems.
- Integrating actuators, sensors, control electronics, and power systems into a small, lightweight structure remains an engineering challenge.
- The proposed hand combines anthropometric modelling, additive manufacturing, and microcontroller-based control in a PLA structure that can be customized quickly.
- Voice recognition, Arduino control, Bluetooth communication, and servo-driven tendon actuation execute coordinated finger flexion and extension through predetermined gestures.
- The design targets simple, low-cost assistive applications such as object manipulation, rehabilitation training, and educational prototyping, while remaining limited to predefined gestures.
2. Methodology
The system translates trained voice commands into predefined finger gestures through a sender–receiver control pipeline, calibrated servo angles, and tendon-driven actuation. Its open-loop design supports repeatable motion but cannot dynamically regulate grip force as object conditions change.
- The control pipeline converts trained voice commands into record numbers, command IDs, and corresponding receiver-side gestures.
- Bluetooth sender–receiver communication separates voice recognition from actuation and provides prompt command transmission with little delay.
- Five separately actuated servo motors support individual-finger and multi-finger movements through gesture-level command abstraction.
- Approximately 90°, 45°, and 0° servo positions correspond to fully open, half-flexed, and fully closed finger states.
- Servo angles were calibrated through incremental rotation and visual observation to obtain clear, repeatable finger positions while avoiding excessive tendon tension.
- The open-loop controller repeats predefined gestures but cannot dynamically adjust grip force, causing slippage and inconsistent grip across objects with different properties.
3. Implementation
The implementation combines a tendon-driven mechanical hand with individually controlled servos, calibrated joint positions, and modular voice-to-Bluetooth command processing. Hardware integration and embedded control support repeatable multi-finger gestures while maintaining stable operation.
- Mechanical Design and Structural Development: The articulated finger structure uses segmented phalanges and rotational joints to support controlled flexion, extension, and simple grasping actions.PLA additive manufacturing was used for lightweight, easily produced prototype components that support rapid design iteration.
- Evaluation and Reliability: Repeated actuation cycles showed no visible tendon slack, breakage, or loss of tension, while the system maintained reliable and repeatable motion.The tendon material remained within a safe operating range for the motor force, although fatigue and long-term durability were not systematically evaluated.
- Actuation Mechanism and Servo Integration: Independent actuators support both individual-finger movement and predefined synchronized multi-finger gestures through dedicated PWM connections.The implementation maps servo motors to fingers and uses programmed control signals for coordinated movement.
- Embedded Software and Control Implementation: The voice-command subsystem identifies trained commands, encodes them as record numbers, and transmits them from a master Bluetooth unit to a receiver that maps them to gestures.The sender and receiver each use an Arduino microcontroller and HC-05 Bluetooth module in a master-slave arrangement.
4. Results
Testing showed smooth, repeatable tendon-driven finger actuation with coordinated and selective gestures, voice-command response, and grasping across varied objects. The open-loop system remained functional but showed limitations with smooth or differently sized objects and small finger-to-finger variations.
- Control response: The system maintained repeatable finger movement and stable operation across repeated actuation cycles.Visual observations and angular-position trajectories indicated smooth control without sudden or irregular movement, although individual finger speeds differed.
- Functional gestures: The hand performed open hand, fist, pinch, and half-grip gestures through synchronized and differential servo actuation.Synchronized movement engaged multiple fingers together, while differential actuation enabled selective finger movement for pinching.
- Finger motion: Servo-driven tendon transmission produced smooth, continuous flexion with coordinated joint motion resembling basic human finger kinematics.Angular trajectories increased gradually, while similar curves indicated coordination; differences in bending were attributed to tendon routing and mechanical tolerances.
- Limitations: Open-loop control caused minor slipping on smooth surfaces and inconsistent grip for objects with different thicknesses or resistance.The system could not dynamically respond to changes in object resistance and surface properties.
- Voice control: Voice commands were recorded, loaded, and translated into open-hand, fist, pinch, and point movements without observable delay.The voice module interacted effectively with the microcontroller, and command-to-motion associations were repeatable across experiments.