Author: Shitalkumar Joshi
The automotive industry is substantially transforming, driven by rapid technological advancements and evolving consumer expectations. From electric vehicles (EVs) to autonomous systems, software-defined vehicles (SDVs), and connected cars, the sector is embracing innovation at an unprecedented pace. Engineering simulation sits at the heart of this revolution, redefining how automakers design, test, and validate vehicles in a fast-changing landscape. Given the significant transformations in automotive product development, simulations are becoming more pervasive, agile, and predictive. AI augmented simulations are emerging as a key enabler, unlocking the full potential of simulations while addressing the ever-evolving challenges in automotive product development.
Technology and Industry Trends
The automotive industry and artificial intelligence (AI) are experiencing disruptive changes unseen in decades. On the automotive side, SDVs, autonomy, and sustainability are the cornerstones of next-generation vehicles. This evolution puts immense pressure on product design engineers. Engineering simulations have always played a crucial role in automotive development, but with the emergence of cyber-physical systems, new simulation methodologies are becoming even more critical.
Simultaneously, AI is transforming, fueled by advances in neural network models and the wide-scale accessibility of AI tools. AI/ML and simulation are distinct technologies that, when intersected, complement each other in compelling ways. In automotive product engineering, AI drives enhanced productivity and introduces new capabilities across all stages of product development, from concept and design to testing, validation, and operation.
With the increasing complexity of software-driven vehicles in electric and autonomous mobility, design engineers face several challenges:
–Design Complexity: The coexistence of New Technology Introduction (NTI) and New Product Introduction (NPI) in the same cycle reduces prior product knowledge, increasing time, cost, and quality risks.
–Testing Requirements: Increased software hardware integration demands more extensive testing scenarios and edge cases to ensure safety and performance, making failures costly.
–Cost Management: Engineers must balance significant new technology introductions (NTI) with budgetary constraints to keep products affordable.
–Safety and Security: In complex cyber-physical vehicles, cybersecurity and functional safety are critical concerns. Stringent regulations must be complied with, in addition to meeting performance and durability goals.
Addressing Challenges with AI-Augmented Simulation
Automakers are integrating AI-augmented engineering simulations throughout the product development cycle to navigate these complexities. Recent advances in AI algorithms and simulation technologies enable simulations in domains previously considered infeasible. AI-augmented simulations enhance traditional simulations by learning from historical test data and synthetic simulations, making them more predictive and efficient. This approach enables faster design optimizations and facilitates early-stage concept verification, helping automotive companies analyze thousands of design iterations more quickly and accurately.
AI-Augmented Simulation in Different Stages of Product Development
-Early-Stage Design Validation: AI-augmented simulation at the early stage enables predicting physics performance in minutes and exchanging ideas with an AI copilot to explore uncharted design spaces.
Most automotive developments are built on previous-generation designs. AI can leverage historical design data to develop learning models that provide real-time insights, enabling engineers to make informed decisions early in the design process.
The learning model empowers designers to assess the performance of new designs within minutes without requiring an expert to build and execute a complete simulation—a combination of generative design and topology optimization enabling engineers to explore and iterate multiple CAD-ready
solutions. AI Augmented simulation approach at the concept stage accelerates the “shift- left” journey, allowing for earlier identification and resolution of potential issues.
* Detailed Design with AI:
AI is enhancing traditional Computer-Aided Engineering (CAE) by optimizing, making it possible to considerably reduce the time needed to run the analyses. For example, crash-test analysis —one of the most data-heavy exercises in automotive design— the traditional simulation can take several days. When AI is combined with simulation, crashes can be analyzed thousands of times faster.
Such a cutting-edge approach will have profound implications for the auto industry, AI-driven optimization techniques and metamodels help engineers derive deeper insights, improving design efficiency and sustainability without compromising speed.
* Enabling Virtual Validation: Virtual testing and validation can reduce development time and costs by up to 40%. AI-powered hybrid analytics combine machine learning, physics-based simulation, and testing data to improve model predictability and real-world accuracy.
* Predictive Models for Operations: AI-powered digital twins continuously learn from telemetry data, evolving into accurate replicas of real-world assets. This enables more precise prognostics, diagnostics, and operational optimizations, which can be implemented through over-the-air updates.
AI in Automotive Applications
AI-augmented simulation is already transforming key automotive applications:
–External Aerodynamics: Generative AI automotive aerodynamic assessment enables a continuous representation of pressure and velocity fields with accuracy of drag prediction ~ 0.5% and speed of simulation increased by 100-1000X across all design phases. This capability dramatically eases optimization between “styling needs” and “aerodynamics efficiency.”
–Road Test for Durability: AI reduces the computational burden of long-duration durability tests by extrapolating large datasets from shorter test simulations.
–Heat Sink Design for Electronics Cooling: AI-augmented simulations evaluate new heat sink topologies almost instantaneously, optimizing thermal management.
–Battery Aging and SoH Estimation: AI-driven simulations predict battery health by combining machine learning with accelerated battery aging tests, improving service scheduling and reliability.
-Traction Machine Control Calibration: AI-based electrical models streamline traction motor controller calibration, reducing testing time by more than 40%
-ADAS Sensor Simulation: AI-powered physics-based simulations generate training data for advanced driver-assistance system (ADAS) algorithms, accelerating their development and improving safety.
The Road Ahead
AI-augmented simulation is revolutionizing automotive engineering, enhancing design capabilities, accelerating the transition to electric and autonomous vehicles, and ensuring compliance with stringent regulatory standards. As digital transformation reshapes the automotive sector, AI-driven simulations will be pivotal in driving innovation and efficiency. By enabling faster, more precise, and cost-effective development processes, AI-augmented simulations provide automakers with a strategic advantage in an increasingly competitive market, empowering them to bring next- generation vehicles to market confidently.
About Author
Shitalkumar Joshi, Senior Director Application Engineering, Ansys. He has 22 years of experience providing thought leadership and cutting-edge solution to organizations in adopting digital transformation journey. He has field engineering exposure working across the industry segments: automotive, aerospace, semiconductor, electronics. he has been involved in deployment of programs around process and practices through tracking timelines, resources, and budget.
