Resource allocation using PDDL
Machines (robots) are present almost in every part of the world today and human lives have benefited from the development of these machines. Some machines are endowed with Artificial Intelligence (AI) having their own decision making capability to react according to any abrupt change in the environm...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/77613 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Machines (robots) are present almost in every part of the world today and human lives have benefited from the development of these machines. Some machines are endowed with Artificial Intelligence (AI) having their own decision making capability to react according to any abrupt change in the environment which they operate in. Most of these intelligent agents are able to carry out tasks in a predefined sequence or strategy which can be described using Planning Domain Definition Language (PDDL) or other AI planning and action languages. PDDL is being explored in this project to enable resource allocation for a multi-agent system. The intelligent agents in the multi-agent system are each required to perform a specific task to achieve the overall goal of the system. The ultimate aim of resource allocation, optimization, is about several heterogeneous intelligent agents achieving a specific goal using the shortest possible time and overcoming any constraints faced during execution. The PDDL was implemented at the top level whereby the functional behavior of the intelligent agents is described. The agents deployed during mission execution uses image-based recognition to detect targets. During image processing, one of the primary functions in identification of targets is edge detection. In this project, we will study the different edge detection techniques to determine its suitability in the context of mission execution and planning. We experimented with several edge detection techniques using digitized images. The results of our experimentation will serve as a basis for targets identification in the mission execution platform configured to demonstrate the cooperative mission planning framework. |
|---|