Summary
Self-driving cars can be designed in different ways. One way uses a single deep learning system that directly turns what the car sees into actions, while another breaks the process into separate parts like seeing, thinking, and controlling. The authors compare these two approaches, pointing out that end-to-end systems are simpler but harder to understand and need lots of data. Modular systems are easier to interpret and can be finely tuned but can be complex to build and expand. They also introduce a guide to help choose the best design for different situations and suggest combining both methods for better results.
What this means in practice
- •For autonomous vehicle developers: Select suitable design approaches using a four-dimensional framework based on safety, data resources, and operating environment.
- •For industrial robotics teams: Consider hybrid architectures combining end-to-end and modular elements to enhance adaptability and interpretability in robotics control systems.
A survey. It maps existing work.
Abstract
Autonomous driving systems have become a central focus of intelligent transportation research, with End-to-End Learning and Modular Architectures offering two prominent design paradigms for their implementation. E2E Learning uses deep learning algorithms to map raw sensory inputs directly to driving actuators, providing a streamlined and adaptable solution. while Modular Architectures employ a pipeline-based approach, dividing the system into distinct subsystems for perception, cognition, planning, and control. This paper presents a comprehensive comparative analysis of these paradigms, focusing on their strengths, limitations, and trade-offs to provide insights into their suitability for various autonomous driving applications. The study evaluates key factors such as interpretability, scalability, robustness, and real-world applicability. While End-to-End Learning emphasizes simplicity and adaptability in dynamic environments, it lacks transparency and is highly dependent on large datasets. Conversely, Modular Architectures offer superior interpretability and task-specific optimization, but face challenges related to integration complexity and scalability. To address these limitations, hybrid approaches that combine the strengths of both paradigms have emerged, offering a promising direction for overcoming these challenges. Beyond this comparative synthesis, following work proposes a Four-Dimensional Architecture Selection Framework, comprising twelve binary criteria across safety, operating environment, data/computational resources, and deployment context, and validate it against ten published autonomous driving systems, correctly recommending 7/10 deployed architectures. This work synthesizes existing literature to highlight key trade-offs between the paradigms and identifies hybrid architectures as a promising direction for future research.