Loading...

An Emotion-Based Intelligent Guidance Planning for a Flapping Vehicle

Haeri, Amin | 2017

381 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49842 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Pourtakdoust, Hossein
  7. Abstract:
  8. In this research, an emotion-based intelligent guidance planning for a flapping vehicle is proposed. It utilizes different sciences including artificial intelligence, neural, fuzzy and psychology accompanied by flight dynamics and modeling. The proposed system enables the robot to make emotion, learn and make decision based on its given personality. The decision making is related to the path planning and how moving from the current position to the designed one. In other words, the robot, at each moment, knows what to do and how to do. Therefore, guidance commands of the flying robot are as outputs of the system. In the present research, the control system is assumed ideal. The introduced intelligent guidance system is divided into two phases: offline and online. First, the robot is trained for a specific mission by the learning subsystem offline. Then, the both emotion modeling subsystem and decision-making subsystem are added to the main system in the online phase. In this phase, the robot has the ability to choose optimal action and reach to the position by utilizing action-to-commands subsystem. In order to design and implement the system, a new mission is defined: how to train an aerial robot to surreptitiously reach the target; i.e. the robot should learn how to use static obstacles to be hidden from dynamic obstacles and touch the target. A large number of simulations reveal that the flying robot has the ability to succeed in over 80 percent of the random missions. In should be also mentioned that the computational load of the system is low
  9. Keywords:
  10. Flapping Wing ; Autonomous Robot ; Decision Making ; Learning ; Emotion ; Intelligent Guidance

 Digital Object List

 Bookmark

No TOC