Automotive Test & Simulation - May 2026
Machine learning drives design space exploration…how simulation is accelerating ADAS innovation…what OEMs must do to spin the AI flywheel. Read all about it in this compendium of articles from the editors of Automotive Engineering magazine.
Don't have an account?
Overview
The May 2026 Automotive Test & Simulation Special Report explores cutting-edge advancements in automotive engineering powered by AI, machine learning (ML), and advanced simulation, highlighting how these technologies transform design, diagnostics, and development workflows to address industry challenges.
A key focus is the integration of probabilistic ML, notably through Secondmind’s active learning approach, which enables engineers to efficiently explore complex design spaces by intelligently selecting simulation runs based on prediction uncertainties. This method breaks from traditional brute-force simulation by reducing costly runs from thousands to hundreds while increasing confidence in design feasibility across variables like temperature, load, and ambient conditions. It supports a cultural shift from intuition-driven decisions toward evidence-based, quantified trade-offs in cost, weight, and performance, accelerating agreement among engineering teams and management.
The report also details Neural Concept’s AI-native workflow, emphasizing AI-augmented engineers or "quantitative designers" who generate and evaluate vast design variant landscapes rather than refining single points. This reshape of engineering processes leverages an intelligence layer that works atop existing digital tools (CAD, CAE, PLM), unlocking rapid iteration, system-level insights, and collaborative decision-making. By collapsing traditional iterative cycles and scaling data-driven insights, automotive companies can significantly improve R&D agility, reduce development cycles by up to 30%, and save millions on vehicle programs.
Battery diagnostics feature prominently via Energsoft’s software-defined impedance-native framework that moves beyond traditional telemetry-based analytics. By unifying Electrochemical Impedance Spectroscopy (EIS), partial impedance data, pulse response, and dQ/dV measurements within a physics-constrained model, the approach enables earlier mechanistic interpretation of degradation and safety risks using existing data without new hardware. This framework enhances early fault detection and predictive maintenance across battery lifecycles, addressing the limitations of indirect voltage/current telemetry signals.
Advanced driver assistance systems (ADAS) development benefits from Dassault Systèmes’ MODSIM platform that facilitates traceability, regulatory compliance, and integrated simulation of radar, electromagnetic interference, sensor fusion, and large-scale connectivity models (e.g., 6G in smart cities). Virtual testing complements physical validation by simulating extreme and edge conditions to ensure robustness in complex real-world environments.
Overall, the report underscores how combining AI, ML, and simulation transforms automotive engineering from fragmented, intuition-heavy workflows to connected, intelligent, and scalable processes. This empowers engineers to manage rising system complexity and regulatory demands, innovate faster, and turn simulation from a bottleneck into a competitive advantage in the rapidly evolving automotive landscape.
