Driving · 2026-10-02
End-to-End Learning vs. Modular Architectures: Comparative Insights into Autonomous Driving Systems
Kartik B. Kapse
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
The authors compare end-to-end driving (raw sensory inputs to control commands) with modular driving pipelines (perception, cognition/decision, planning, control), summarizing trade-offs in interpretability, scalability, robustness, and real-world suitability. They further propose a Four-Dimensional Architecture Selection Framework (12 binary criteria) and validate it retrospectively on ten deployed autonomous driving systems, recommending the correct architecture for 7/10 cases (4 exact, 3 partial).
Why it matters
Choosing an autonomous driving architecture impacts safety, explainability, robustness, and deployment feasibility. By synthesizing trade-offs between end-to-end and modular paradigms and providing a criteria-based framework validated against deployed systems, the paper aims to help practitioners select an architecture type for a given context.
Method
- Provides a comparative synthesis of end-to-end learning vs modular architectures for autonomous driving across factors like interpretability, scalability, robustness, and real-world applicability.
- Defines a Four-Dimensional Architecture Selection Framework (twelve binary criteria across safety/regulatory compliance, operating environment, data/computational resources, and deployment/operational context) mapping to Modular, end-to-end Imitation Learning, end-to-end Reinforcement Learning, or Hybrid recommendations.
- Validates the framework retrospectively against ten representative deployed autonomous driving systems (1989–2023) by comparing predicted architecture to each system’s actual deployed design.
Limitation
The paper does not follow a formal systematic review protocol (e.g., PRISMA); it adopts a narrative synthesis approach appropriate for comparing heterogeneous work across three decades, sensor modalities, and evaluation benchmarks.
Abstract (from arXiv)
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.
Related papers
- A Survey on End-to-End Autonomous Driving Training from the Perspectives of Data, Strategy, and Platform
- Beyond Planning: Differentiable Control-in-the-loop End-to-End Autonomous Driving
- Think Densely, Act Sparsely: Latent Expert Cognitive Chains for Vision-Language-Action Autonomous Driving
- AD-E2E-JEPA: A Joint-Embedding Predictive Architecture For End-to-End Autonomous Driving
- EvoAD: Self-Evolving Autonomous Driving via Latent World-Model Guided Post-hoc Reflection
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: 2026-10-01 · Indexed by Pub-AI: 2026-10-02 UTC
End-to-End Learning vs. Modular Architectures: Comparative Insights into Autonomous Driving Systems
Preprint posted: 2026-10-01 · Original record ↗
- arxiv
- 2610.01746
Explicit source versions (1)
- v1 · 2026-10-01
- Preprint available
Source date: 2026-10-01 · First detected: 2026-10-25 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · atom:published (official metadata)
Abstract · arxiv
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.