/// TEST SIMULATION \\\
challenges in complex tests
Monolith Artificial Intelligence Software creates new possibilities in specifying complex crash, aerodynamic and environmental tests
A
collaboration with the BMW Group has seen artificial intelligence software provider Monolith accelerate the de-
velopment of the manufacturer’s vehicle range by applying its AI platform. By training Monolith’s self-learning models with the com- pany’s wide-reaching engineering test data, engineers can now use AI to solve previously impossible physics challenges and instantly predict the performance of highly complex systems like crash and aerodynamics tests. The BMW Group crash test engineering
team began working with Monolith in 2019 via the BMW Startup Garage to explore the po- tential of using AI to predict the force on a passenger’s tibia during a crash. Current crash development uses thousands of simu- lations as well as physical tests to capture performance. Even with sophisticated mod- elling, owing to the complexity of the physics underpinning crash dynamics, results require substantial engineering know-how to cali- brate for real-world behaviour.
Moreover, physical crash tests can only be
conducted in later stages of development when the design is mature enough to create physical prototypes. Exploring a more effi- cient method, the BMW Group collaborated with Monolith to see whether AI could predict crash performance and, importantly, do so substantially earlier in the vehicle develop- ment process.
Self-learning models \\\ Using Monolith, BMW Group engineers built
self-learning models using the wealth of their existing crash data and were able to accu- rately predict the force on the tibia for a range of different crash types without doing physi- cal crashes. Additionally, the accuracy of the self-learning models will continue to improve as more data becomes available and the plat- form is further embedded into the engineer- ing workflow. This approach now means engineers can optimise crash performance earlier in the design process and reduce de-
pendence on time-intensive and costly test- ing while making historical data infinitely more valuable. According to Dr Richard Ahlfeld, CEO and
founder of Monolith, when the intractable physics of a complex vehicle system means it cannot be truly solved via simulation, AI and self-learning models can fill the gap to in- stantly understand and predict vehicle per- formance. This offers engineers a tremendous new tool to do less testing and more learning from their data by reducing the number of re- quired simulations and physical tests while critically making existing data more valuable. “We are excited to see how BMW Group en-
gineers are using pioneering technology like Monolith to reduce the cost and time of prod- uct development as they develop the next generation of vehicles,” he adds.
Wider test applications for AI \\\ The Monolith platform has been developed
Artificial Intelligence software platform is providing BMW engineers with a dashboard on which test outcomes can be evaluated for different parameter values
with a focus on user experience by automo- tive experts and data scientists to ensure seamless integration with existing engineer- ing processes. As a result, as soon as the soft- ware is implemented, domain experts quickly begin gaining valuable insights and saving time, as well as the chance to explore even wider opportunities. In this respect, BMW is expanding its use of Monolith into more engineering functions across R&D that generate vast amounts of data from different sources, such as aerody- namics and advanced driver-assist systems (ADAS). There is the opportunity to use Artifi- cial Intelligence to predict outcomes in chal- lenging applications such as noise, vibration and harshness (NVH) testing and other envi- ronmental tests like climatic testing. In all these cases, the physics is complex and the number of possible test scenarios is large. “The combination of engineering expertise and machine learning can provide a significant competitive edge and provide our customers with the means to create world-class products more efficiently.” Ahlfeld concludes. C&VT
2022 /// Climatic & Vibration Testing \\\ 7
AI solves physics
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40
