ACCURATE PREDICTIONS OF WILDFIRE SPREAD ARE CRITICAL FOR EFFECTIVE WILDFIRE MANAGEMENT TO SUPPORT DECISION MAKINGS OF FIRE MANAGERS AND TO ENSURE SAFETY OF FIREFIGHTERS. HOWEVER, THE LACK OF REAL TIME WILDFIRE AND WIND DATA, BOTH OF WHICH CHANGE IN SPACE AND TIME, MAKES IT DIFFICULT TO ACHIEVE OPERATIONAL WILDFIRE SPREAD PREDICTION. UNMANNED AIRCRAFT SYSTEM (UAS) IS EMERGING IN MANY CIVILIAN APPLICATIONS AND SHOWS GREAT POTENTIAL IN WILDFIRE MANAGEMENT. THIS PROJECT AIMS TO DEVELOP AND EVALUATE A COLLABORATIVE HUMAN-UAS WILDFIRE SPREAD PREDICTION AND SITUATIONAL-AWARENESS SYSTEM FOR WILDFIRE MANAGEMENT. THE UASS WILL WORK SIDE-BY-SIDE WITH FIRE MANAGERS AND GROUND FIREFIGHTERS TO PERFORM COLLABORATIVE TASKS. THIS NEW PARADIGM BRINGS NEW RESEARCH CHALLENGES FROM MULTIPLE ASPECTS. FIRST AND FOREMOST, THE UASS MUST ACHIEVE SUFFICIENT AUTONOMY IN THEIR MISSION SO THAT THEY CAN AUTONOMOUSLY COLLECT THE MOST USEFUL INFORMATION IN DYNAMIC WILDFIRE ENVIRONMENTS. BESIDES WILDFIRE SENSING, THEUASS ALSO NEED TO PAY CLOSE ATTENTION TO FIREFIGHTERS' SAFETY BY MONITORING THEIR VICINITY. THE SECOND CHALLENGE IS ASSOCIATED WITH EFFECTIVE TEAMING AND COLLABORATION BETWEEN HUMANS (FIRE MANAGERS AND FIREFIGHTERS) AND UASS. IN PARTICULAR, THERE IS A NEED FOR HUMANS TO INTERACT WITH AND DIRECT UASS' AUTONOMY BASED ON THEIR DOMAIN KNOWLEDGE AND EXPERT OPINIONS FOR MORE EFFECTIVE WILDFIRE MANAGEMENT. TO ADDRESS THESE CHALLENGES, THIS PROJECT WILL INCLUDES FOUR TASKS: (1) FIRE SENSING AND WIND ESTIMATION USING A TEAM OF UASS TO ENABLE DATA-DRIVEN WILDFIRE SPREAD PREDICTION, (2) UAS COORDINATION AND PATH PLANNING ALGORITHMS GOVERNING UAS AUTONOMY TO SENSE DYNAMIC WILDFIRES WHILE MONITORING FIREFIGHTERS' SAFETY RISK, (3) TEAMED HUMAN-UAS COLLABORATION, INCLUDING HUMAN-DIRECTED AUTONOMY AND A HUMAN-UAS INTERACTION INTERFACE TO SUPPORT HUMAN AWARENESS OF UAS OPERATION, AND (4) EVALUATION OF THE PROPOSED RESEARCH BY FLYING A TEAM OF UASS OVER PRESCRIBED FIRES.THIS PROJECT HAS THEPOTENTIAL TO TRANSFORM WILDFIRE MANAGEMENT BY ENABLING OPERATIONAL WILDFIRE SPREAD PREDICTION AND SITUATION AWARENESS FOR FIREFIGHTERS THROUGH TEAMED HUMAN-UASS COLLABORATION. USING UASS TO SENSE FIRE CHARACTERISTICS AND WIND PARAMETERS WILL FILL THE CRITICAL GAP OF REAL TIME DATA COLLECTION AND DATA ASSIMILATION FOR OPERATIONAL WILDFIRE SPREAD PREDICTION. THE MULTI-UAS AUTONOMY ALGORITHMS ALLOW UASS TO EFFECTIVELY COLLECT THE MOST USEFUL INFORMATION ABOUT DYNAMIC WILDFIRES AND TO MONITOR THE SAFETY OF FIREFIGHTERS AND OTHER PEOPLE ON THE GROUND. THE APPROACH OF HUMAN-DIRECTED AUTONOMY SUPPORTS HUMANS IN-THE-LOOP TO OPTIONALLY DIRECT UAS TEAMS TO CERTAIN LOCATIONS AND TASKS FOR EFFECTIVE HUMAN-UAS COLLABORATION. BESIDES WILDFIRE MANAGEMENT, THIS RESEARCH WILL ALSO BENEFIT OTHER EMERGENCY RESPONSE APPLICATIONS IN WHICH HUMANS AND AUTONOMOUS ROBOTS INCREASINGLY WORK TOGETHER. THE PIS WILL DEVELOP NEW AND UNIQUE EDUCATION AND OUTREACH PROGRAMS, INCLUDING A WILDFIRE-UAS FIELD TRIP PROGRAMAND AN ANNUAL OUTREACH WORKSHOP SERIES TO PROVIDE INTERDISCIPLINARY TRAINING TO UNDERGRADUATE/GRADUATE STUDENTS AND TO OUTREACH TO BROADER COMMUNITIES AND THE GENERAL PUBLIC.