Specialised mathematical computing software provider MathWorks is presenting its technologies and solutions at the 2018 FISITA World Automotive Congress. The company’s major products include MATLAB and Simulink, which support data analysis and simulation. At the FISITA Congress, MathWorks is showcasing technologies around Deep Learning, Advanced Driver Assistance Systems (ADAS), motor control leveraging Model-Based Design, and achieving ISO 26262 functional safety certification.
At the event, we caught up with Wensi Jin, Automotive Industry Manager, MathWorks and R Vijayalayan, Application Engineering Manager, MathWorks India Pvt Ltd. Jin is responsible for strategic planning and technology rollout. His focus is to foster industry adoption of Model-Based Design and MathWorks tools. He holds a degree in Electrical Engineering from the University of Texas at Austin. Meanwhile, Vijayalayan manages the Control Design Application Engineering in MathWorks India and specialises in the field of Model-Based Calibration, Electrification, Full Vehicle Modeling and Model-Based Design. He holds a Bachelor’s Degree from Manonmaniam Sundaranar University and a Master’s Degree in Control and Instrumentation from IIT Madras.
ATR _ How has been the journey of MathWorks in India?
Wensi Jin (WJ) _ MathWorks is celebrating its tenth anniversary in India this year, and the local team has grown well over the years. The operations initially began with customer support, which is currently carried out at four offices across the country. This was followed up by the setting up of two R&D centres that work along with the customer teams as well as develop tools for the company’s global customers. While one R&D centre focusses on simulation, test and verification, the other was set up to carry out work around automated driving.
What is your view about FISITA 2018 and what do you expect from this event?
WJ _ FISITA brings engineers from a lot of different disciplines and provides a platform, where we can see how they are all being brought together through diverse sessions. Another good outcome is that the automotive industry has become more global in nature, and this event brings together engineers from various countries and assures their varied requirements and helps in talking to the entire automotive supply chain that includes OEMS, suppliers, services providers and so on. You see the whole automotive ecosystem being reflected here.
Please tell us about the disruptive technology trends being seen in the automotive industry in India.
WJ _ We call these megatrends, whether they would be disruptive or not is to be seen. But these are definitely forces within the industry, as well as outside the industry are looking to leverage. For us these megatrends are electrification, autonomy and connectivity, from MathWorks’ point of view. There are three main commonalities for the company in addressing these trends, with the first being the use of software and algorithms. MathWorks plays prominently in the software space, with the software content increasing year over year. It is fair to say that software is the differentiator for a car today, and while customers don’t buy software directly, a lot of the features they buy are based on the software.
The second area is data. Software has traditionally been based on logic, but nowadays you see the combination of logic-based software as well as data-driven software (Machine Learning and Deep Learning). The final criterion is the fact that technology is getting more complex, which is leading to growth in the drive towards simulation. There is a mindset shift towards simulation, where it helps rule out basic dimensional problems during development.
How is MathWorks trying to provide solutions for the present requirements of the automotive industry?
WJ _ The company has always been in the space of the mentioned megatrends, especially in the off-road sector where automation has been around for some time. That was actually the first application of MathWorks tools for autonomy development. We look at the structure of an autonomous system, which has sensors, followed by the fusion system that brings multiple sensors together, and finally culminating in planning and decision. All these different disciplines have been implemented with MathWorks tools. We have a different toolbox for different applications, and that has been established years ago. This is where we are in terms of supporting autonomy. Engineers can easily open a toolbox and begin doing perception development for autonomous driving.
However, one new idea being carried out over the past five years is called Reference Applications, which is use as a workflow to tackle industry problems. An example is that there is tremendous demand for autonomous driving engineers, which exceeds supply. There is a shift of people working on areas like powertrain, body or chassis systems toward automated driving. These engineers need to come up to speed as quickly as possible to learn the workflow and look at end-applications. This causes many new teams to come together, and the use of Reference Applications can speed up learning and actual development.
Another new solution from the company is in the tools and reference for predictive maintenance. The company also launched its Vehicle Dynamics concept late last year, and that is to provide collateral vehicle dynamics. The Vehicle Dynamics concept is more like creating a virtual test ground to test and verify ADAS functions.
Could you give us an Indian perspective of the MathWorks story?
R Vijayalayan (RV) _ The Indian scenario consists of both domestic companies as well as global MNCs representing OEMs and suppliers. It is being seen that both these types of players are focusing on the mentioned megatrends. However, we particularly see domestic OEMs moving towards electrification and emission norms due to the regulations set by the government. The technologies of simulation and data analysis from MathWorks are raising interest from customers, so that they can be used to improve productivity, reduce testing times and frontload all the development activities. It is being seen that domestic OEMs want to enhance their in-house competencies for development activities of future mobility trends. The company is also working to bridge the industry-academia connect by helping enhancing the curriculum to enable students to apply theory in practical form.