DLP - Evolutionary Mobile Robots using Computational Intelligence Techniques
Evolutionary robots, like autonomous artificial organisms, automatically develop their own skills by interaction with the environment. This talk will focus on evolutionary locomotion control of mobile robots using computational intelligence techniques, including fuzzy systems and evolutionary computation. First, the basic concept of evolutionary fuzzy systems (EFSs) will be introduced. Next, for wheeled robots, an obstacle boundary following behavior learned through EFSs will be introduced. Evolutionary fuzzy control of a single wheeled robot and multiple wheeled robots cooperatively carrying an object through multi-objective evolutionary computation algorithms for obstacle boundary following will be introduced. Then, to boost the learning efficiency of multi-objective EFSs in this application, the technique of reinforcement neural fuzzy surrogate-assisted learning will be given. Finally, navigation of a single and multiple cooperative wheeled robots in unknown environments will be presented.
Date and Time
Location
Hosts
Registration
- Date: 14 Sep 2023
- Time: 10:00 AM to 11:30 AM
- All times are (UTC+08:00) Kuala Lumpur
- Add Event to Calendar
- Starts 01 September 2023 06:17 PM
- Ends 13 September 2023 06:17 PM
- All times are (UTC+08:00) Kuala Lumpur
- No Admission Charge
Speakers
Prof Chia-Feng Juang
Evolutionary Mobile Robots using Computational Intelligence Techniques
Evolutionary robots, like autonomous artificial organisms, automatically develop their own skills by interaction with the environment. This talk will focus on evolutionary locomotion control of mobile robots using computational intelligence techniques, including fuzzy systems and evolutionary computation. First, the basic concept of evolutionary fuzzy systems (EFSs) will be introduced. Next, for wheeled robots, an obstacle boundary following behavior learned through EFSs will be introduced. Evolutionary fuzzy control of a single wheeled robot and multiple wheeled robots cooperatively carrying an object through multi-objective evolutionary computation algorithms for obstacle boundary following will be introduced. Then, to boost the learning efficiency of multi-objective EFSs in this application, the technique of reinforcement neural fuzzy surrogate-assisted learning will be given. Finally, navigation of a single and multiple cooperative wheeled robots in unknown environments will be presented.
Biography: