MathWorks:Enabling OEMs to see tomorrow today

Leading global software provider and developer MathWorks is focusing on four major global trends in the automobile industry. The US-based software company has its programs ready for the future of automotive that include automated driving and electric vehicles.

It is also concentrating on predictive maintenance and functional safety standards for software.

Speaking to Autocar Professional, Vijayalayan R, Vijayalayan R, Manager, Automotive Industry Field Application Engineering Team, MathWorks India says, ”From an automotive perspective, there are four major trends that are driving the industry globally. They are electrification, automated driving, data analytics or predictive maintenance and functional safety standards. These are the key driving factors for our development and we develop our products to suit the needs of the industry. Electrification and automated driving are hot megatrends not only in India but across the globe. So, there is a big need for developing our products quickly."
The software company releases its updated software and products every six months, which is two releases in a year. It constantly keeps adding capabilities to the existing software giving more value to its customers. Major products include MATLAB and Simulink, which support data analysis and simulation. 

Keeping all these in mind, the company increased the quality of the software and has a product called Model-Based Design, which is widely embraced by many automotive OEMs, for a long time. Model-Based Design is a system model is at the centre of the development process, from requirements development, through design, implementation, and testing. This emphasises early verification and improves the quality of the product. 
“There are three different products that help the trends in the industry. The Powertrain Blockset is mainly focused on full vehicle simulations. When people want to do a quick architectural level simulation or system-level simulation of electric vehicles or hybrid electric vehicles, they do not want to build it from scratch. So, we have the reference designs from which they can take and add parameters with their data and quickly use it for simulation models," says Vijayalayan. 

MathWorks India's Vijayalayan R: "Three of our products — Powertrain Blockset, System Composer and Automated Driving Toolbox — can help industry ride trends."

He adds, "Also, we have recently introduced something for system engineering, which helps in defining the system architecture itself. Called the System Composer, one can allocate requirements while refining an architecture model that can then be designed and simulated in Simulink. The third one is the Automated Driving Toolbox. This provides algorithms and tools for designing, simulating, and testing ADAS and autonomous driving systems. You can design and test vision and lidar perception systems, as well as sensor fusion, path planning, and vehicle controllers. All these will help our customers for the major trends in the market. “

Global connect
MathWorks brings out one common product which has use globally. Though there will be support from the team of engineers sitting in India, there is no separate product that is region-specific. If required, this team will go to its customer place and assist them. For example, the Simulink Platform was used by an electric two-wheeler manufacturer in India and also by a well-known electric carmaker in the US, customised based on their specific requirements. 

“With the help of our technical team support and also consulting engineering, they were able to adopt the Model-Based Design in adapting the tool to their particular needs. So, no major customisation was done in the tool. But, the partnership was to understand how to use the tools for their particular needs,” explains Vijayalayan.

According to him, all new products will be driven by trends like autonomous systems, electrification and artificial intelligence. In the autonomous system space, MathWorks is also looking at the products supporting aerial drones. 

According to him, the company has been using AI for a long time. Every time you remove a human and add an algorithm in the loop, it becomes smarter. AI has been part of MathWorks software as algorithms for a long time. They have a toolbox that has deep learning capabilities, which really helps in incorporating all the deep learning algorithms and come up with the right accurate models.

Making the most of data  insight for predictive maintenance
“There is also a huge number of data that is available and collected from the system. This data helps us to get insights and use it as predictive maintenance. So, artificial intelligence can be seen in multiple places from algorithm development and also far from data analytics, extracting insights and then using it for making decisions for smart manufacturing, predictive maintenance. These are widely used, but we are scaling it up. So, when it is said AI, it has to be a combination of intelligence in the model which can be integrated with engineering systems. We have the strength and we give it as one platform together,” states Vijayalayan. 

Nowadays, simulation plays a major role in designing and making a vehicle. But simulating a vehicle in a system and bringing it alive without a physical prototype, is still a myth in most of the cases. Vijayalayan says it is not possible. Pointing out a disadvantage of building a physical prototype, he said, "Prototypes will be a challenge and will take up a huge amount of time and cost. And, there are certain scenarios that cannot be predicted. "

He continues, “But at the same time, it will be difficult to replace full physical prototype testing. Maybe, the number of prototypes built can be reduced. With more dependence on simulation, one can reduce the number of hours spent on vehicle testing. And, there is a possibility that many can come up with newer ideas even before a physical model is made. People can quickly conceptualise their ideas and be ready with their algorithms. On the whole, simulation brings down the development time and the testing time,” he concludes.