Journal of Applied Data Sciences
Vol 2, No 2: MAY 2021

Training Autonomous Vehicles in Carla model using Augmented Random Search Algorithm

Riyanto Riyanto (Universitas Amikom Purwokerto, Indonesia)
Abdul Azis (Universitas Amikom Purwokerto, Indonesia)
Tarwoto Tarwoto (Universitas Amikom Purwokerto, Indonesia)
Wei Li Deng (Southeast University, Nanjing, China)

Article Info

Publish Date
23 Apr 2021


CARLA is an open source simulator for autonomous driving research. CARLA has been developed from scratch to support the development, training and validation of autonomous driving systems. In addition to open source code and protocols, CARLA provides open digital assets (urban layouts, buildings, vehicles) that are created for this purpose and can be used freely. We use CARLA to study the performance of Augmented Random Search (ARS) to autonomous driving: a classic modular pipeline, an end-to-end model trained via imitation learning, and an end-to-end model trained via reinforcement learning. Test the ability of the Augmented Random Search (ARS) algorithm to train driverless cars on data collected from the front cameras per car. In this study, a framework that can be used to train driverless car policy using ARS in Carla will be built. Although effective policies were not achieved after the first round of training, many insights on how to improve these outcomes in the future have been obtained.

Copyrights © 2021

Journal Info





Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management


One of the current hot topics in science is data: how can datasets be used in scientific and scholarly research in a more reliable, citable and accountable way? Data is of paramount importance to scientific progress, yet most research data remains private. Enhancing the transparency of the processes ...