group
What campus are you from?
Daytona Beach
Authors' Class Standing
Carmen Dimario, Senior Ihan Ventura, Senior Austin Rollfing, Senior Michael Ulivarri, Senior Daniel Minks, Senior
Lead Presenter's Name
Carmen Dimario
Faculty Mentor Name
Dr. Akbas
Abstract
ERAU RoboRacer's autonomous racing presents unique challenges for real-time decision-making, requiring vehicles to operate at the limits of performance while maintaining safety in dynamic environments. It aims to create a 1:10-scale autonomous race car that integrates perception, planning, and control to compete in RoboRacer competitions while serving as a research platform for high-speed autonomy. The system fuses data from LiDAR, cameras, and IMU sensors to perform simultaneous localization and mapping, detect obstacles and competitors, and plan optimal racing lines for both time-trial and head-to-head racing. Control is achieved through a hybrid approach combining pure pursuit, PID velocity control, and model predictive control, all running onboard an NVIDIA Jetson embedded compute module. Safety is integrated throughout the architecture via startup diagnostics, continuous sensor health monitoring, and automatic emergency stop capabilities. Beyond competition readiness, this work contributes a maintainable ROS2-based development framework including a simulation environment for algorithm validation before hardware deployment. This platform enables investigation of multi-sensor fusion, safe control under uncertainty, and the practical tradeoffs inherent in physical autonomous racing systems.
Did this research project receive funding support from the Office of Undergraduate Research.
No
ERAU RoboRacer
ERAU RoboRacer's autonomous racing presents unique challenges for real-time decision-making, requiring vehicles to operate at the limits of performance while maintaining safety in dynamic environments. It aims to create a 1:10-scale autonomous race car that integrates perception, planning, and control to compete in RoboRacer competitions while serving as a research platform for high-speed autonomy. The system fuses data from LiDAR, cameras, and IMU sensors to perform simultaneous localization and mapping, detect obstacles and competitors, and plan optimal racing lines for both time-trial and head-to-head racing. Control is achieved through a hybrid approach combining pure pursuit, PID velocity control, and model predictive control, all running onboard an NVIDIA Jetson embedded compute module. Safety is integrated throughout the architecture via startup diagnostics, continuous sensor health monitoring, and automatic emergency stop capabilities. Beyond competition readiness, this work contributes a maintainable ROS2-based development framework including a simulation environment for algorithm validation before hardware deployment. This platform enables investigation of multi-sensor fusion, safe control under uncertainty, and the practical tradeoffs inherent in physical autonomous racing systems.