Unlocking Edge AI in Software-Defined Vehicles: Balancing model complexity, compute resources, and data access

The potential of Edge AI in automotive goes beyond ADAS and automated driving. AI-driven optimizations in driving efficiency, in-cabin personalization, predictive maintenance, and enhanced safety require intelligent processing at the vehicle edge. However, integrating AI into current vehicle electrical/electronic architectures (EEAs) presents significant challenges. This paper examines how AI models can run on general-purpose ECUs, gateways, and domain controllers while balancing model complexity, compute resources, and data availability. We explore trade-offs, architectural choices, and the need for software configurability to ensure AI models evolve with new data and vehicle parameters. We conclude by providing a glimpse into Sonatus’ latest AI product that encapsulates a holistic approach optimized for vehicle platforms—spanning cloud-based model development, optimization, and validation to in-vehicle inference—that unlocks the potential of Edge AI and SDVs for the automotive industry. Sonatus is a leading software solution supplier that empowers vehicle manufacturers to accelerate the shift to SDVs as platforms for innovation rapidly. Our solutions help build flexible EEAs and add software configurability to vehicles in areas that include network management, data collection and management, in-vehicle software orchestration, and OTA updates. Our solutions, including a precise and intelligent data collection solution, have been deployed on millions of vehicles, supporting hundreds of product innovations and analytics use cases. Flexible and intelligent access to vehicle data is critical to building and deploying in-vehicle AI models to improve vehicle performance and deliver unique customer experiences. In this paper, we will discuss how the software infrastructure elements stated above, as well as targeted data acquisition, are critical to the success of in-vehicle AI.