London-based AI software company Monolith uses the AI platform to significantly reduce the cost of developing and testing new cars.
Monolith’s software uses self-learning models to instantly predict the outcome of complex vehicle dynamics systems, reducing the need for physical testing and simulation. According to the company, this approach is dramatic at every stage of the automotive development process, from initial design to design iteration, validation, and production that now requires iterative, time-consuming, and costly testing and simulation. Accelerate to. Monolith states that the platform requires fewer physical prototypes and road tests, making product validation safer and more sustainable.
“Optimizing a system, or finding a new solution based on 10 years of historical data, is like giving engineers 10 years of experience instantly. That’s the power of AI. For corporate data. By unleashing the stored expertise, we strengthen our expertise in personal subjects, “said Dr. Joel Henry, Principal Engineer at Monolith.
So far, automotive companies have used a combination of life-like virtual simulations and physics tests during vehicle development. At each design iteration, simulation solves the physics that underpins system modeling. This is a very difficult and computationally intensive process. Virtual simulations help reduce the number of physical tests required, but can limit the accuracy and fidelity of the results. Therefore, many physical tests are still required to calibrate and validate virtual results and understand their performance under operating conditions that cannot be simulated.
“Today, automotive companies are spending billions of dollars on developing electrical architectures and software features to win the competition for electricity, sharing, and autonomous mobility, which allows them to invest in research and development budgets and products in other areas. Timeline is squeezed and puts a lot of pressure on the engineering team working to develop higher quality vehicle hardware systems with less time and less resources, “said Dr. Richard Ahlfield, CEO and Founder of Monolith. ..
Monolith has spent six years developing a platform for merging virtual and physical test data to train high-precision AI self-learning models. The model then predicts the performance of the system by understanding how it behaves, instead of solving the complex physics of the system or performing physical tests.
“Monolith was founded to empower engineers with AI to instantly solve even the most unwieldy physics problems. What is this in hundreds of complex simulations? We know that it resonates especially with auto engineers who are struggling to optimize hundreds of often conflicting criteria. Engineers who need hours or days to resolve have virtual test limits. We’re still frustrated with the significant amount of physics testing needed to make up for it. At the same time, the data created by the process represents a huge opportunity when used in AI, “Ahlfield said. increase.
https://www.electronicsworld.co.uk/london-company-monolith-brings-artificial-intelligence-to-automotive-development/33900/ London company Monolith brings artificial intelligence to car development – Electronics World