A gamified simulator and physical platform for self-driving algorithm training and validation

TL;DR

This paper presents a gamified simulator and physical platform for training and validating self-driving algorithms, enabling high-quality data collection, transfer learning, and safe deployment in real-world scenarios.

Contribution

It introduces a novel gamified simulator for data collection and a physical test platform that together enhance self-driving algorithm training and validation.

Findings

Synthetic data enables CNN-based vehicle control

Model transfer from simulation to physical platform is successful

The platform supports diverse data collection and rapid testing

Abstract

We identify the need for a gamified self-driving simulator where game mechanics encourage high-quality data capture, and design and apply such a simulator to collecting lane-following training data. The resulting synthetic data enables a Convolutional Neural Network (CNN) to drive an in-game vehicle. We simultaneously develop a physical test platform based on a radio-controlled vehicle and the Robotic Operating System (ROS) and successfully transfer the simulation-trained model to the physical domain without modification. The cross-platform simulator facilitates unsupervised crowdsourcing, helping to collect diverse data emulating complex, dynamic environment data, infrequent events, and edge cases. The physical platform provides a low-cost solution for validating simulation-trained models or enabling rapid transfer learning, thereby improving the safety and resilience of self-driving algorithms.