Bridging the gap: A low-cost, ROS-based testbed for rapid prototyping of autonomous outdoor navigation

The development and validation of autonomous vehicle (AV) systems remain complex, costly, and high-risk endeavors. While simulations suffer from a “reality gap” and full-scale vehicle tests are prohibitively expensive, the lack of intermediate platforms creates a gap for agile, real-world algorithm prototyping. This paper presents the design and implementation of an affordable, small-scale car-like testbed that serves as a crucial bridge between simulation and full-scale deployment. Our platform is built using commercial off-the-shelf (COTS) components, including a Raspberry Pi, a low-cost GPS/IMU, and a standard RC car chassis, all orchestrated by a modular ROS-based software architecture. To demonstrate its effectiveness, we implement a complete Guidance, Navigation, and Control (GNC) stack, first verifying its design in simulation with realistic sensor and actuator models, followed by a successful demonstration of autonomous outdoor waypoint navigation in real-world tests. The complete platform provides an accessible, safe, and flexible tool for researchers and educators to rapidly develop, test, and benchmark autonomous driving algorithms.