Multi-Agent Control Using Fuzzy Logic
| Main Author: | |
|---|---|
| Language: | English |
| Published: |
University of Cincinnati / OhioLINK
2015
|
| Subjects: | |
| Online Access: | http://rave.ohiolink.edu/etdc/view?acc_num=ucin1447688633 |
| id |
ndltd-OhioLink-oai-etd.ohiolink.edu-ucin1447688633 |
|---|---|
| record_format |
oai_dc |
| spelling |
ndltd-OhioLink-oai-etd.ohiolink.edu-ucin14476886332021-08-03T06:33:43Z Multi-Agent Control Using Fuzzy Logic Cook, Brandon M. Aerospace Materials Fuzzy Logic UAS Traffic Management Unmanned Aerial Systems Intelligent Systems UAV Collision Avoidnace Multi-Agent Control In the coming years, operations in low altitude airspace will vastly increase as the capabilities and applications of Unmanned Aerial Systems (UAS) continue to multiply. Therefore, solutions to managing vehicles in highly congested airspace must be explored. In this study, an intelligent systems approach was used to help mitigate the risk of collision between aircraft in uncontrolled airspace using a UAS Traffic Management (UTM) System. To test the effectiveness of this system, a three-dimensional environment was created using MATLAB to simulate a fully autonomous heterogeneous fleet of UAS attempting to accomplish a variety of realistic missions, including precision agriculture, package delivery services, natural resource monitoring, and disaster management. Main research challenges include situational awareness, decision making, and multi-agent control in an uncertain, time-critical, spatio-temporal environment. To gain the knowledge, experience, and expertise necessary to solve this large-scale real-world problem, two preliminary research efforts were conducted. First, a simulated gaming platform known as Pong, originally created by ATARI, was used to demonstrate the effectiveness of a fully autonomous team to accomplish a desired task using a cascading Fuzzy system. With this knowledge, a simplified UTM system was developed to test a preliminary design of a fuzzy collision avoidance system. Once complete, this knowledge was used to develop the final UTM system platform capable of using intelligent separation assurance and collision avoidance techniques to mitigate the risk for Near Mid-Air Collisions between aircraft. This fuzzy solution utilizes only current state information and can resolve potential conflicts without knowledge of intruder intent. The collision avoidance system was tested in extreme conditions, including close proximity, high closure rates, and conservative maximum turn rates. In the preliminary homogenous case, the collision avoidance techniques were on average 99.977% successful over a span of nearly 2,485 flight hours. Whereas, in the final UTM platform consisting of heterogeneous agents, the collision avoidance system was on average 99.88% successful over a span of 16,255 flight hours. Lastly, it was found that the techniques employed for separation assurance drastically mitigated the risk for Near Mid-Air Collisions. Comparing the unmitigated and mitigated cases, the number of losses of separation between aircraft reduced from one loss of separation per two flight hours, to one loss of separation per ten flight hours. This mitigated separation assurance platform was successful at preventing a loss of separation 88.47% of the time, over a span of 7,545 flight hours. 2015 English text University of Cincinnati / OhioLINK http://rave.ohiolink.edu/etdc/view?acc_num=ucin1447688633 http://rave.ohiolink.edu/etdc/view?acc_num=ucin1447688633 unrestricted This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws. |
| collection |
NDLTD |
| language |
English |
| sources |
NDLTD |
| topic |
Aerospace Materials Fuzzy Logic UAS Traffic Management Unmanned Aerial Systems Intelligent Systems UAV Collision Avoidnace Multi-Agent Control |
| spellingShingle |
Aerospace Materials Fuzzy Logic UAS Traffic Management Unmanned Aerial Systems Intelligent Systems UAV Collision Avoidnace Multi-Agent Control Cook, Brandon M. Multi-Agent Control Using Fuzzy Logic |
| author |
Cook, Brandon M. |
| author_facet |
Cook, Brandon M. |
| author_sort |
Cook, Brandon M. |
| title |
Multi-Agent Control Using Fuzzy Logic |
| title_short |
Multi-Agent Control Using Fuzzy Logic |
| title_full |
Multi-Agent Control Using Fuzzy Logic |
| title_fullStr |
Multi-Agent Control Using Fuzzy Logic |
| title_full_unstemmed |
Multi-Agent Control Using Fuzzy Logic |
| title_sort |
multi-agent control using fuzzy logic |
| publisher |
University of Cincinnati / OhioLINK |
| publishDate |
2015 |
| url |
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1447688633 |
| work_keys_str_mv |
AT cookbrandonm multiagentcontrolusingfuzzylogic |
| _version_ |
1719439169111982080 |