Research Symposium Program - Individual Details
5th annual Undergraduate Research Symposium, April 17, 2025
Liliana Carlson C- 2 R - 5
BIO
Liliana Carlson is a senior in High School at Ohana Institute from Santa Rosa Beach, FL, with a passion for robotics, engineering, and physics. She has extensive experience in competitive robotics, serving as a FIRST Robotics Competition (FRC) team captain and mentor and leading FTC (FIRST Tech Challenge) projects focused on autonomous programming and drivetrain optimization. Liliana’s expertise includes probabilistic decision-making models, odometry tuning, and swerve drive calibration. In recognition of her leadership, technical skills, and dedication to STEM, she was named a 2024 FRC Dean’s List Finalist.
Beyond robotics, Liliana has a strong academic foundation in physics and mathematics, pursuing advanced coursework in calculus and engineering. She is passionate about pushing the boundaries of automation and robotics and applying innovative problem-solving approaches to real-world challenges.
Maximizing FRC Autonomous Efficiency with Probabilistic Decision-Making Models
Authors: Liliana Carlson, Milinda Jay Stephenson, Ph.DStudent Major: Biomedical Engineering
Mentor: Milinda Jay Stephenson, Ph.D
Mentor's Department: English Mentor's College: Florida State University Panama City Co-Presenters:
Abstract
This paper explores the application of probabilistic decision-making models to enhance the efficiency and reliability of autonomous operation in FIRST Robotics Competition (FRC) robots. Traditional deterministic autonomous programs often struggle with variability in sensor data, drivetrain inconsistencies, and unforeseen obstacles, leading to suboptimal performance. Robots can dynamically adjust their actions based on real-time conditions by integrating probabilistic models, optimizing movement strategies, and increasing scoring efficiency. Drawing from research in autonomous vehicles and Bayesian networks, this study examines how probabilistic frameworks improve adaptability and decision-making in uncertain environments. Key findings demonstrate that probabilistic approaches enhance FRC autonomous strategies by enabling real-time adjustments, reducing error rates, and maximizing competitive performance. The results suggest that future FRC teams can benefit from incorporating probabilistic modeling techniques to develop more robust and flexible autonomous routines.
Keywords: Autonomous, Decision-Making, Robotics, Probabilistic