A Software Architecture for Autonomous Vehicles: Team LRM-B Entry in the First CARLA Autonomous Driving Challenge

Abstract

The objective of the first CARLA autonomous driving challenge was to deploy autonomous driving systems to lead with complex traffic scenarios where all participants faced the same challenging traffic situations. According to the organizers, this competition emerges as a way to democratize and to accelerate the research and development of autonomous vehicles around the world using the CARLA simulator contributing to the development of the autonomous vehicle area. Therefore, this paper presents the architecture design for the navigation of an autonomous vehicle in a simulated urban environment that attempts to commit the least number of traffic infractions, which used as the baseline the original architecture of the platform for autonomous navigation CaRINA 2. Our agent traveled in simulated scenarios for several hours, demonstrating his capabilities, winning three out of the four tracks of the challenge, and being ranked second in the remaining track. Our architecture was made towards meeting the requirements of CARLA Autonomous Driving Challenge and has components for obstacle detection using 3D point clouds, traffic signs detection and classification which employs Convolutional Neural Networks (CNN) and depth information, risk assessment with collision detection using short-term motion prediction, decision-making with Markov Decision Process (MDP), and control using Model Predictive Control (MPC).

Integrating Modular Pipelines with End-to-End Learning: A Hybrid Approach for Robust and Reliable Autonomous Driving Systems

“…In the Track SENSORS, our hybrid CaRINA agent achieved a route completion score of 85.01%, surpassing other autonomous driving methods primarily based on end-to-end learning. We also obtained a high driving score (DS = 35.36) compared to similar approaches (WOR [62], MaRLn [63], NEAT [64], AIM-MT [64], TransFuser [65], CNN-Planner [66], Learning by Cheating [67], CILRS [68], CaRINA 2019 [10]).…”

“…This section introduces our hybrid architecture for mapless autonomous driving, designed to navigate challenging scenarios like the CARLA Leaderboard's SENSORS track. Building upon our modular pipeline described in previous sections, we leverage robust obstacle detection, risk assessment, and decision-making modules while replacing traditional map-based planning with an end-to-end path planner named the CNN-planner [10].…”

“…The 2023 CADCH represents the fifth edition of the autonomous driving competition set in urban simulated environments using the CARLA, featuring diverse urban scenarios. Our involvement spans multiple editions, and we secured victory in the inaugural challenge through a modular and vision-based navigation stack [10]. Therefore, the primary contributions of this paper are the following.…”