**AWARDS ISSUED PRIOR TO JANUARY 20, 2025, WERE FUNDED UNDER PREVIOUS ADMINISTRATIONS AND MAY NOT REFLECT THE PRIORITIES AND POLICIES OF THE CURRENT ADMINISTRATION.** THE US HAS BEEN THE WORLD'S LEADER IN BRINGING INNOVATIVE SOLUTIONS TO MODERN AGRICULTURE AIMED AT IMPROVING THE QUANTITY AND QUALITY OF AGRICULTURAL PRODUCTS, WHILE MINIMIZING ECONOMIC AND ENVIRONMENTAL COSTS.LARGE FARM SIZES HAVE CATAPULTED THE INDUSTRY TO USE BIG MACHINERY, DRIVEN BY THE NEED TO COMPLETE FIELD OPERATIONS IN A TIMELY MANNER WITH MINIMAL LABOR. HOWEVER, LARGER MACHINERY HAS ALSO LED TO LONG-TERM ECONOMIC AND ENVIRONMENTAL CONCERNS RESULTING FROM SOIL COMPACTION, LIMITED ABILITY TO ADDRESS SMALL-SCALE FIELD VARIABILITY, AND REDUCED CROP PRODUCTIVITY.SEVERAL FIELD TASKS, SUCH ASMONITORING CROP HEALTH, WEED SCOUTING ANDCONTROL, ASSESSING SOIL FERTILITY STATUS, AND HARVESTING,ARE STILL TYPICALLY PERFORMED BY HUMAN WORKERS AND REQUIRE TRAINED LABOR. MANY LOW-COST, SMALL ROBOTS WITH EMBEDDED SWARM INTELLIGENCE BRING NOVEL APPROACHES TO ADDRESS THESE CHALLENGES AND ACHIEVE SYSTEM-LEVEL FLEXIBILITY AND IMPROVED ROBUSTNESS. THIS PROJECT WILL ESTABLISH A CONFIGURABLE, ADAPTIVE, AND SCALABLE SWARM (CASS) SYSTEM CONSISTING OF UNMANNED GROUND AND AERIAL ROBOTS FOR COLLABORATIVE SMART AGRICULTURE BY:(1)ESTABLISHING THE OVERALL SYSTEM ARCHITECTURE AND LAY THE TECHNICAL GROUNDWORK;(2)DEVELOPING THE CASS SYSTEM CONSISTING OF THE PHYSICAL HARDWARE PLATFORM, ITS DIGITAL TWIN, AND A USER INTERFACE (UI); AND(3)EVALUATING THE CASS SYSTEM FOR ITS TECHNICAL FUNCTIONALITY, TASK PERFORMANCE, AND USABILITY, THROUGH TESTBED EVALUATIONS AND TWO REAL-WORLD CASE STUDIES.OUR LONG-TERM GOAL OF ESTABLISHING A DEPLOYABLE, EASY-TO-USE CASS SYSTEM THAT CAN SERVE AS A UNIVERSAL PLATFORM FOR BROAD AGRICULTURE APPLICATIONS BY: (A) ESTABLISHING THE TECHNICAL AND THEORETICAL GROUNDWORK FOR THE DEPLOYABLE, SCALABLE SWARM SYSTEM, (B) DEVELOPING AND INTEGRATING CORE COMPONENT TECHNOLOGIES, AND (C) PRODUCING EVIDENCE-BASED KNOWLEDGE VIA COMPREHENSIVE EVALUATIONS. THIS NRI PROJECT SERVES AS A CRITICAL PATHWAY TOWARDS THIS LONG TERM GOAL.UTILIZING A SCALABLE MOBILE AD-HOC NETWORK AND A NOVEL SWARM DECISION MAKING TECHNIQUE BASED ON A PROBABILISTIC FINITE STATE MACHINE MODEL, THE CASS SYSTEM WILL SEAMLESSLY INTEGRATE THE HARDWARE PLATFORM, ITS DIGITAL TWIN, AND THE UI. THE HARDWARE PLATFORM WILL EMPLOY A HETEROGENEOUS TEAM OF CONFIGURABLE GROUND AND AERIAL ROBOTS. THE DIGITAL TWIN WILL SUPPORT I) HIGH-FIDELITY SIMULATIONS OF INDIVIDUAL ROBOTS FOR LOCOMOTION AND INDIVIDUAL SENSOR PERFORMANCE AND II) LOW-FIDELITY SIMULATIONS OF THE ENTIRE SWARM AND LARGER TASK TEAMS FOR SWARM BEHAVIOR AND COLLECTIVE TASK PERFORMANCE ONDIGITAL ENVIRONMENTAL MODELS CREATED FOR EACH TARGET APPLICATION.TWO REAL-WORLD CASE STUDIES WILL BE USED FOR EVALUATION: (STUDY I) FIELD SCOUTING AND WEED MANAGEMENT AND (STUDY II) DETERMINING LOCATIONS AND BEHAVIOR OF GRAZING BEEF COWS.