AUTOFRAME -- A Software-driven Integration Framework for Automotive Systems

2025-03-06

AI Summary

To address the challenges of escalating system complexity and low iteration efficiency inherent in distributed architectures for Software-Defined Vehicles (SDVs), this paper proposes a centralized-computing-based hardware-software integration framework. The framework introduces a novel, modular, scalable, and secure deployment mechanism that uniquely integrates a Hardware Abstraction Layer (HAL) with a dynamic software deployment engine. It enables cross-platform adaptive integration and over-the-air (OTA) updates of core autonomous driving functions—including lane detection, motion planning, and vehicle control. Validated via simulation across diverse autonomous driving scenarios, the framework significantly reduces software iteration cycles, enhances system maintainability, and demonstrates feasibility for deployment in real in-vehicle environments.

Technology Category

• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Application Domains: Mobility, Driving & Flight

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

The evolution of automotive technologies towards more integrated and sophisticated systems requires a shift from traditional distributed architectures to centralized vehicle architectures. This work presents a novel framework that addresses the increasing complexity of Software Defined Vehicles (SDV) through a centralized approach that optimizes software and hardware integration. Our approach introduces a scalable, modular, and secure automotive deployment framework that leverages a hardware abstraction layer and dynamic software deployment capabilities to meet the growing demands of the industry. The framework supports centralized computing of vehicle functions, making software development more dynamic and easier to update and upgrade. We demonstrate the capabilities of our framework by implementing it in a simulated environment where it effectively handles several automotive operations such as lane detection, motion planning, and vehicle control. Our results highlight the framework's potential to facilitate the development and maintenance of future vehicles, emphasizing its adaptability to different hardware configurations and its readiness for real-world applications. This work lays the foundation for further exploration of robust, scalable, and secure SDV systems, setting a new standard for future automotive architectures.

Problem

• Centralized architecture for Software Defined Vehicles
• Scalable and secure automotive deployment framework
• Dynamic software and hardware integration optimization

Innovation

• Centralized architecture optimizes software-hardware integration
• Scalable, modular framework with hardware abstraction layer
• Dynamic software deployment supports centralized vehicle functions