BEGIN:VCALENDAR VERSION:2.0 METHOD:PUBLISH PRODID:-//Missouri State University/Calendar of Events//EN CALSCALE:GREGORIAN X-WR-TIMEZONE:America/Chicago BEGIN:VTIMEZONE TZID:America/Chicago BEGIN:DAYLIGHT TZOFFSETFROM:-0600 TZOFFSETTO:-0500 DTSTART:20070311T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU TZNAME:CDT END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0500 TZOFFSETTO:-0600 DTSTART:20071104T020000 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU TZNAME:CST END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:a91d5320-f11d-42f9-b148-b9e6899dd673.192463@calendar.missouristate.edu CREATED:20181031T151518Z LAST-MODIFIED:20181031T151518Z LOCATION:Cheek Hall 0151 SUMMARY:An Adaptive Memory Based Reinforcement Learning Controller DESCRIPTION:Thesis Defense - Keith Cissell\n\n\nAbstract:\n\n\nRecently\, the use of autonomous robots for exploration has drastically expanded--la rgely due to innovations in both hardware technology and the development of new artificial intelligence (AI) methods. The wide variety of robotic agents and operating environments has led to the creation of many unique control strategies that cater to each specific agent and their goal withi n an environment. Most control strategies are single purpose\, meaning th ey are built from the ground up for each given operation. Here we present a single\, reinforcement learning control solution for autonomous explor ation intended to work across multiple agent types\, goals\, and environm ents. Our solution includes a memory of past actions and rewards to effic iently analyze an agent’s current state when planning future actions. The agent’s objective is to safely navigate an environment and collect data to achieve a defined goal. The control solution is first compared with ra ndom and heuristic control schemas. To test the controller for adaptabili ty\, the controller is next subjected to changes in the agent’s sensors\, environments\, and goals. Control strategies are compared by examining  goal completion rates\, the number of actions taken\, and the agent’s rem aining health and energy at the end of a simulation. Results indicate tha t our newly developed control strategy is adaptable to new situations. A reinforcement-learning based controller\, such as the one presented in th is research\, could help provide a universal solution for controlling aut onomous robots in the field of exploration. \n\n\n  X-ALT-DESC;FMTTYPE=text/html:<html><head><title></title></head><body><p>Th esis Defense - Keith Cissell</p>\n<p>Abstract:</p>\n<p><span lang="en-US" >Recently\, the use of autonomous robots for exploration has drastically expanded--largely due to innovations in both hardware technology and the development of new artificial intelligence (AI) methods. The wide variety of robotic agents and operating environments has led to the creation of many unique control strategies that cater to each specific agent and thei r goal within an environment. Most control strategies are single purpose\ , meaning they are built from the ground up for each given operation. Her e we present a single\, reinforcement learning control solution for auton omous exploration intended to work across multiple agent types\, goals\, and environments. Our solution includes a memory of past actions and rewa rds to efficiently analyze an agent’s current state when planning future actions. The agent’s objective is to safely navigate an environment and c ollect data to achieve a defined goal. The control solution is first comp ared with random and heuristic control schemas. To test the controller fo r adaptability\, the controller is next subjected to changes in the agent ’s sensors\, environments\, and goals. Control strategies are compared by examining<span>&nbsp\; </span>goal completion rates\, the number of acti ons taken\, and the agent’s remaining health and energy at the end of a s imulation. Results indicate that our newly developed control strategy is adaptable to new situations. A reinforcement-learning based controller\, such as the one presented in this research\, could help provide a univers al solution for controlling autonomous robots in the field of exploration .</span><span lang="en-US"> </span></p>\n<p><span lang="en-US">&nbsp\;</s pan></p></body></html> DTSTART;TZID=America/Chicago:20181112T110000 DTEND;TZID=America/Chicago:20181112T120000 SEQUENCE:0 URL: CATEGORIES:Current Students,Faculty,Staff END:VEVENT END:VCALENDAR