Development of Pedestrian Artificial Intelligence Utilizing Unreal Engine 4 Graphic Engine
Fares Abu-Abed1, Alexey Khabarov2
1Fares Abu-Abed, Faculty of International Academic Cooperation / Tver State Technical University, Tver, Russian Federation.
2Alexey Khabarov, Faculty of Information Technologies/ Tver State Technical University, Tver, Russian Federation.
Manuscript received on 14 April 2019 | Revised Manuscript received on 18 May 2019 | Manuscript published on 30 May 2019 | PP: 639-642 | Volume-8 Issue-1, May 2019 | Retrieval Number: A7288058119/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Paper demonstrates the implementation of AI for pedestrian simulation in the driving simulator by means of Unreal Engine 4 graphic engine and VR-technologies. The authors review the operation principles of behavior trees and their components. The paper describes the internal structure of a pedestrian AI class and methods of implementing object detection in the field of view and audibility utilizing Unreal Engine 4. The authors give the example of using Environment Query System included in the engine AI system and display the result of executing its queries, which used in the process of the virtual pedestrian behavior tree simulation. Since in the developed driving simulator it is necessary to achieve a high frequency of frame changes and low demands on the resources of the computer system, the article suggests an optimal solution for simulating a large number of pedestrians in a virtual city. The article shows the behavior tree and represents operation algorithms of the pedestrian AI’s basic components in block diagrams with annotations.
Index Terms: Unreal Engine 4, Environment Query System, Behavior Tree, Artificial Intelligence, Driving Simulator
Scope of the Article: Artificial Intelligence and Machine Learning