Hand-held haptic interfaces for multi-degree-of-freedom haptic guidance
Abstract/Contents
- Abstract
- Haptic guidance can be effective for learning or performing many physical tasks involving touch and motor control. World-grounded haptic interfaces use rigid links that provide net forces and torques to physically push a user's hand or body through intended movements. However, for many applications, a world-grounded haptic interface is not suitable because it limits the user's workspace and freedom of motion. This dissertation addresses the design and use of holdable haptic devices for providing movement guidance to a user's hands, with potential application to medical procedures, sports training, robot teleoperation, and navigation. Because a holdable device does not ground reaction forces to an external surface, there are limitations on the forces and torques it can apply to a user's hand. This thesis focuses on overcoming these limitations through novel designs and control techniques. First, we designed a unique haptic device that uses two control moment gyroscopes (CMGs) to isolate desired moment axes and generate clear, directional cues that users can easily identify. By reorienting the CMG gimbals between cues, moment pulses are generated about any 3D axis without transient off-axis components. Actuating the CMGs with an asymmetric pulse pattern generates sequential haptic cues for closed-loop orientation guidance. Torque measurements and user studies validated the device's ability to provide effective guidance. Second, we designed two two-fingered, holdable haptic devices that guide a user by applying lateral fingertip deformation. User studies demonstrated that grounding through a handle can enable intuitive motion cues in multiple degrees of freedom. With one device design, we showed that both translation and rotation cues can be generated by the same device action by displacing fingertips along a curve and varying the displacement magnitude. With a second device design, we used a pair of two-degree-of-freedom end-effectors to expand the possible direction cues applied to users' fingers. We measured movements in response to open-loop guidance cues to validate that each direction is clear and distinguishable. Another study showed that users can follow desired paths using closed-loop cues from the device. Because guidance with these devices depends on the perception and interpretation of haptic cues, we developed several methods for modeling and controlling haptic guidance that can be individually tailored to each user. For the CMG device, we developed Gaussian process models and used approximate dynamic programming to produce control policies for the gyroscopic device that account for a user's perceptual biases, kinematic constraints, and variability. For the fingertip device, we generated linear state-space models based on each user's recorded movements. These models identified the differences in their perception and dynamics and allowed us to develop personalized linear quadratic and model predictive controllers for trajectory following. Even in the absence of vision and net forces, users were able to follow guidance from our devices to reach targets and follow trajectories in three dimensions. Results demonstrate the promise of holdable haptic devices for guidance during manual tasks. We found that limitations due to the lack of grounding for forces and torques, such as saturation and user differences, can be improved through controller design. Systems that require mobility, large workspaces, and unrestricted hand movement may be able to incorporate haptic guidance using the techniques presented in this dissertation
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2020; ©2020 |
| Publication date | 2020; 2020 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Walker, Julie Marie |
|---|---|
| Degree supervisor | Okamura, Allison |
| Thesis advisor | Okamura, Allison |
| Thesis advisor | Follmer, Sean |
| Thesis advisor | Kochenderfer, Mykel J, 1980- |
| Degree committee member | Follmer, Sean |
| Degree committee member | Kochenderfer, Mykel J, 1980- |
| Associated with | Stanford University, Department of Mechanical Engineering. |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Julie M. Walker |
|---|---|
| Note | Submitted to the Department of Mechanical Engineering |
| Thesis | Thesis Ph.D. Stanford University 2020 |
| Location | electronic resource |
Access conditions
- Copyright
- © 2020 by Julie Marie Walker
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...