Research outputs per year
Research outputs per year
Project Details
Description
This PCFC Project investigated the control using neural networks as feed forward estimators and as feedback controllers and using in series neural networks.
That was compared to the control of robots using bounded, smoothed, incremental and stepping variable structure control.
Then neural networks were added to the inner feedback path to improve on the use of pre-trained feed forward estimators.
Finally, fuzzy logic was used to approximate the performance test data of a DC-motor connected to a dynamometer to see if the se of fuzzy logic could improve on low level actuator control.
That was compared to the control of robots using bounded, smoothed, incremental and stepping variable structure control.
Then neural networks were added to the inner feedback path to improve on the use of pre-trained feed forward estimators.
Finally, fuzzy logic was used to approximate the performance test data of a DC-motor connected to a dynamometer to see if the se of fuzzy logic could improve on low level actuator control.
Layperson's description
Control of a robot using neural networks and fuzzy logic.
Key findings
Investigated control of a robot using:
- Neural networks as feed forward estimators.
- Neural networks as feedback controllers.
- An in series neural network.
- Bounded, smoothed, incremental and stepping variable structure control.
- Neural networks added to the inner feedback path.
- Fuzzy logic.
- Neural networks as feed forward estimators.
- Neural networks as feedback controllers.
- An in series neural network.
- Bounded, smoothed, incremental and stepping variable structure control.
- Neural networks added to the inner feedback path.
- Fuzzy logic.
| Short title | PCFC Project concerning Robot control using AI and fuzzy logic |
|---|---|
| Status | Finished |
| Effective start/end date | 24/03/90 → 16/03/93 |
Keywords
- Robotics
- Control Engineering
- Neural networks
- Fuzzy logic
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.
-
The addition of neural networks to the inner feedback path in order to improve on the use of pre-trained feed forward estimators
Sanders, D., Haynes, B. P., Tewkesbury, G. & Stott, I. J., Aug 1996, In: Mathematics and Computers in Simulation. 41, 5-6, p. 461-472Research output: Contribution to journal › Article › peer-review
-
Control of a robot using an in series neural network
Sanders, D., Haynes, B. & Sietas, C., 1994, Proceedings E U R I S C O N ' 9 4 European Robotics and Intelligent Systems Conference. Vol. 3 (stream 3). Advanced Manufacturing and Automation Research Centre, p. 1787 - 1795Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
-
Control of a robot using bounded, smoothed, incremental and stepping variable structure control
Sanders, D., Hewer, N. & Tewkesbury, G., 1994, Proceedings E U R I S C O N ' 9 4 European Robotics and Intelligent Systems Conference. Vol. 2 (stream B). Advanced Manufacturing and Automation Research Centre, p. 1131 - 1147Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
-
European Robotics and Intelligent Systems Conference
Sanders, D. (Invited speaker)
25 Aug 1994Activity: Participating in or organising an event types › Participation in conference
-
Control of a robot using bounded, smoothed, incremental and stepping variable structure control
Sanders, D. (Speaker)
24 Aug 1994Activity: Talk or presentation types › Oral presentation
-
Control of a robot using neural networks as feed forward estimators and as feedback controllers
Sanders, D. (Speaker)
23 Aug 1994Activity: Talk or presentation types › Oral presentation