YITONG SUN

Force Feedback Glove

2021

"Force Feedback Glove combines shape-memory alloy actuation, electrorheological damping, and hand sensing to create a bidirectional haptic interface for touching virtual objects."

Force Feedback Glove

Project Description

Force Feedback Glove is a wearable haptic interface designed to simulate the sensation of touching and interacting with virtual objects. The project investigates how compact actuators, variable-resistance materials, sensing systems, and AI-based modeling can be combined to create a bidirectional interface between the human hand and a virtual environment.

The glove is built around three major subsystems: an electronic control and sensing layer, a nickel-titanium shape-memory alloy force-feedback structure, and an electrorheological fluid damping system. These components work together to generate both fast tactile resistance and slower, more continuous force modulation.

The electronic system is responsible for spatial sensing, actuator control, and communication with the virtual environment. An IMU-based sensing system provides orientation and motion feedback, while the control circuit drives the force-feedback elements and reads sensor data from the glove.

The second subsystem is a specially designed nickel-titanium alloy shape-memory spring structure that extends from the fingertips to the back of the hand. Nickel-titanium alloy, also known as NiTi or Nitinol, can contract rapidly when electrically heated. In this glove, the contraction of the shape-memory alloy spring creates a fast mechanical response, allowing the system to simulate sudden contact, impact, or the initial resistance of a virtual object.

The third subsystem is a custom electrorheological fluid damping module located on the back of the hand. Electrorheological fluid changes its flow resistance when exposed to different electric fields. By applying different voltages, the system can adjust the damping resistance experienced during finger movement. This allows the glove to produce a more gradual, linear, and controllable resistance profile.

By combining the fast response of shape-memory alloy actuation with the adjustable resistance of electrorheological damping, the glove aims to create a more convincing tactile experience for virtual object interaction.

Background

Most virtual reality systems are visually immersive but physically limited. Users can see and manipulate virtual objects, but their hands often receive little or no physical feedback. This disconnect reduces the sense of presence and limits the realism of virtual interaction.

Traditional haptic gloves often rely on motors, tendons, brakes, pneumatic actuators, or rigid exoskeletons. These approaches can generate force feedback, but they often increase weight, mechanical complexity, noise, or response delay. A wearable glove needs to be lightweight, responsive, compact, and capable of producing multiple types of force sensations.

This project explores an alternative haptic strategy: instead of using a single actuator type, the glove combines fast contraction and variable damping to simulate different layers of touch feedback.

The shape-memory alloy provides quick force response, while the electrorheological fluid provides controllable resistance over time. This hybrid approach is intended to better match the complexity of real touch, where contact is not only a single impact but also a continuous interaction involving pressure, resistance, friction, and motion constraint.

Technical Method

The glove is organized into three functional layers.

The first layer is the electronic control system. This circuit manages spatial sensing, actuator driving, sensor acquisition, and communication with the virtual system. The sensing system captures motion and orientation information from the hand, while the controller drives both the shape-memory alloy elements and the electrorheological fluid damping module.

The second layer is the NiTi shape-memory alloy spring structure. The spring-like form is designed to fit along the finger-to-hand mechanical path. When current is applied, the nickel-titanium alloy heats and contracts, generating a pulling or resisting force. This rapid contraction is used to simulate immediate haptic events, such as touching the surface of a virtual object, encountering a boundary, or feeling a sudden change in resistance.

The third layer is the electrorheological fluid damping system. This system is positioned at the back of the hand and is designed to provide adjustable resistance during finger movement. By changing the applied voltage, the controller changes the apparent viscosity or flow resistance of the fluid. This enables a more continuous and linear form of force feedback, allowing the glove to simulate different material properties or object stiffness levels.

The two force-feedback mechanisms are complementary. The shape-memory alloy responds quickly and provides a strong initial sensation, while the electrorheological fluid provides slower, tunable damping. Together, they create a layered haptic response that is more nuanced than a simple on/off actuator.

Bidirectional Interaction and Position Feedback

In addition to outputting force feedback to the user, the glove is designed to collect information from the hand and feed it back into the virtual system. By measuring the damping position, the contraction state of the NiTi alloy, and motion data from the sensing system, the glove can estimate finger posture and spatial position.

This data can be processed through AI-based modeling to infer the user's finger configuration. The inferred finger position can then be sent back to the virtual environment, allowing the system to update the digital hand model in real time.

This creates a bidirectional interaction loop:

  • The virtual system detects contact with a virtual object.
  • The glove generates corresponding force feedback through NiTi contraction and ER-fluid damping.
  • The glove senses the user's finger motion and actuator states.
  • AI-based modeling estimates finger position and interaction state.
  • The updated hand position is fed back into the virtual system.

Through this loop, the glove functions not only as an output device for haptic sensation, but also as an input device for hand tracking and interaction modeling.

Significance

The Force Feedback Glove proposes a hybrid haptic architecture for more realistic virtual touch. By combining shape-memory alloy actuation and electrorheological damping, the glove addresses two important requirements in haptic feedback: fast response and continuous force modulation.

The project is significant in several areas:

  • Virtual reality and augmented reality interaction.
  • Wearable haptic interfaces.
  • Human-computer interaction.
  • Soft robotics and smart materials.
  • Tactile simulation for virtual object manipulation.
  • Bidirectional sensing and feedback systems.

The glove demonstrates how smart materials can be used to reduce mechanical complexity while increasing the richness of haptic sensation. Instead of relying entirely on motors or rigid mechanical brakes, the system uses material properties themselves as part of the feedback mechanism.

This approach has potential applications in immersive VR, remote robotic control, rehabilitation training, virtual prototyping, simulation-based education, and digital art experiences where tactile realism is essential.

Future Development

Future versions of the Force Feedback Glove could improve response precision, comfort, and integration with virtual environments. The shape-memory alloy system could be optimized for faster cooling, lower power consumption, and more precise contraction control. The electrorheological damping system could be further refined to improve linearity, durability, and miniaturization.

Additional sensors could be integrated to improve finger-position estimation, including flex sensors, magnetic encoders, pressure sensors, or optical tracking markers. The AI model could also be expanded to learn user-specific hand motion patterns, improving accuracy over time.

A future system could combine haptic feedback, gesture recognition, and real-time virtual physics simulation into a fully interactive wearable platform. With continued development, this glove could become a lightweight and adaptive interface for experiencing digital objects not only visually, but physically.