By navigating our site, you agree to allow us to use cookies, in accordance with our Privacy Policy.

Engineers Don’t Need to Become Data Scientists to Achieve Success in AI

R VijayalayanWith the need for efficiency improvement and meeting the shrinking timelines, there’s a big need for virtualization of the electrified vehicle development and testing using simulation emphasizes R Vijayalayan, Manager, Automotive Industry Field Application Engineering Team, MathWorks India Pvt. Ltd. During an extensive e-interaction with Niloy from BISinfotech the veteran pans across on how AI is solving complex engineering projects, the trends shaping the Indian auto sector, their leadership in Model-Based Design and challenges for Indian Auto sector when it comes to innovating high-tier autonomous segment. Much more in the edited excerpts below.

Q How has been MathWorks supporting complex engineering projects especially when it comes to solving AI challenges?

Engineers working on AI projects often expect to spend a large percentage of their time developing and fine-tuning AI models. Though modeling is an important step in the workflow, the model is not the end of the journey.

MathWorks believes that engineers should consider four steps for a complete AI-driven workflow as it is critical to unearth any issues early on for success in AI implementation. The four-step process consists of Data Preparation, AI Modeling, Simulation and Test and Deployment.

Engineers don’t need to become data scientists to achieve success in AI. MathWorks has designed tools, and apps to help them build and integrate AI into their workflows. The engineers bring in the domain knowledge rather than spending inordinate amounts of time becoming data scientists. They can use apps to quickly try out different approaches and apply their domain expertise to prepare the data. If it’s not feasible to identify features in the data, they can use deep learning, which identifies features as part of the training process. Deep learning requires lots of data, but they can use transfer learning to extend an existing network to work with the data they have. Finally, they can deploy the model as part of a complete AI system on an embedded device.

Q Pivotal focus of MathWorks in the Indian auto sector and if you can elaborate more about your specific offerings in this dramatically changing auto sector?

The first major digital transformation in the auto industry that happened two decades ago put embedded controls in everything, from power windows to door locks. There was a need to fold in software expertise and combine domain knowledge of auto experts with people who make software.

Currently, the automotive industry is going through a second digital transformation focusing on developing a software-defined vehicle.

The automotive industry embraced the Model-Based Design to address the challenges caused due to growing software and system complexity. The systematic reuse of models is a basic principle of Model-Based Design, where models form a digital thread connecting development, design optimization, code generation, and verification and validation. This digital thread does not need to be limited to the development process; it can be extended to deploy systems in operation when design models are reused as digital twins.

Today, MATLAB and Simulink are acknowledged as helping engineers work on automotive projects based on the increase of electronics in vehicles, including focusing on four key trends: electrification, automated driving, data analytics, predictive maintenance, and functional safety standards.

Q Electrification is dominating the future of the auto industry. What has been MathWorks role in helping innovations and ushering customers towards a sustainable electrified future?

Electrified vehicle development requires one to address many challenges in parallel such as Total vehicle efficiency, New engineering capability – power electronics, high-voltage battery, motors, embedded software quality and transition from prototyping work style to production mindset and processes. MathWorks helps engineers tackle these challenges through a unified design environment for system engineering, software engineering, and data science.

With the need for efficiency improvement and meeting the shrinking timelines, there’s a big need for virtualization of the electrified vehicle development and testing using simulation. MathWorks provides vehicle model templates to lower the barrier to start design. Engineers can use these models for design tradeoff analysis and component sizing, control parameter optimization, and hardware-in-the-loop simulations. This approach coupled with data analytics enables engineers to perform virtual testing or calibration which results in saving of time and cost. With the increase in software content in today’s electric vehicles, companies are also turning toward virtual vehicles to test their software as soon as possible.

Q How has been MathWorks transforming the Software Development Paradigm addressing the unique need of customers and industry?

Model-Based Design extends agile principles to the development of systems that include physical components as well as software. From requirements capture, system architecture, and component design, to implementation, verification, test, and deployment – Model-Based Design spans the entire development cycle.

Simulink from MathWorks integrates with Jenkins™, Jira, Git™, GitHub®, and other tools. This lets the teams use models instead of documents to collaborate across teams, integrate their development in continuous integration, build, and test processes and scale development using a combination of desktop and cloud resources.

Twice a year, MathWorks provides a thoroughly tested new release that includes, on average, 500 new features across all products, plus enhancements to existing features and performance improvements. MathWorks runs one million automated software tests per day to ensure the quality and compatibility of new code.

Q ADAS and Autonomous are shaping the future of the auto industry. Your focus on these applications?

The focus of MathWorks is to provide the tools to enable the Automated Driving engineers to gain insight into real-world behaviour by visualizing, replaying, and analyzing vehicle data. Engineers can reduce vehicle testing by modeling and simulating scenarios that are difficult to repeat or are too dangerous to test in the real world. Finally, engineers can verify the functionality of the embedded software by integrating and testing code with the same simulated scenarios. This approach enables engineers to continuously increase confidence in their design along the development process.

MathWorks Products such as MATLAB®, Simulink®, and RoadRunner advance the design of automated driving perception, planning, and control systems.

Q Challenges specific to India when it comes to advancing towards complex technologies like Autonomous in the high-tier segment?

As per the market research and different studies, some of the key challenges that are observed when it comes to advancing towards Autonomous driving in India are the infrastructure readiness in adopting autonomous vehicle practices, workforce mobility and safety certification.

However, with a focus on pedestrian and driver safety, we will see increasing acceptance of the ADAS and automated driving systems in today’s vehicles. Also, we see that there’s an increasing focus from the off-road vehicles and farm vehicles segment on developing fully or semi-autonomous vehicles. We observe that an increasing number of technical services companies, Global in-house technical centres and startups companies are focusing their R&D work in automated driving. But the key challenge before the industry is the availability of skilled expertise. This brings out the need for ramping up skills in testing, simulation, system engineering and bridging the gap between industry and academia through industry-ready curriculum and project-based learning.

  • Support for Startups and Incubators

We are also observing that there are quite a few startups working in the automated driving space. MathWorks supports incubators and startups worldwide with sponsor and discounted benefits. We enable them in reaching early-stage milestones fast with help of our products.

  • Support for Student Competitions

MathWorks prepares and supports the next generation of scientists and engineers with software, training, and mentoring to tackle the same technical issues as professional engineers. We partner with BAJA SAE India, Formula Bharat and runs MathWorks Minidrone competitions supporting student teams with industry-standard tools and mentoring where they can apply classroom theory to competition problems. This helps students gain industry-ready skills.

Q Given 2021, key announcements of MathWorks empowering the Indian Auto sector?

MathWorks has invested in helping engineers and scientists in AUTO industry augment their domain knowledge with our solutions and resources. MathWorks follows a twice-yearly general release schedule. Each general release synchronizes the full MATLAB® and Simulink® product families, delivering new features and bug fixes for existing products and, when available, new products.

For the automotive industry, RoadRunner is an interactive editor that lets you design 3D scenes for simulating and testing automated driving systems. The new RoadRunner Scene Builder product, part of the RoadRunner product family, imports and automatically synthesizes 3D road models from HERE HD Live Map road data.

Powertrain Blockset™ provides fully assembled reference application models of automotive powertrains, including gasoline, diesel, hybrid, and electric systems. It includes a component library for simulating engine subsystems, transmission assemblies, traction motors, battery packs, and controller models. Powertrain Blockset also includes a dynamometer model for virtual testing.

Vehicle Dynamics Blockset now includes the ability to implement 6DOF trailers and vehicles with three axles and Simulink 3D blocks that offer the ability to visualize tractors and trailers in the Unreal Engine 3D environment. Lidar Toolbox is a new product that provides algorithms, functions, and apps for designing, analyzing, and testing lidar processing systems.

Radar Toolbox includes algorithms and tools for designing, simulating, analyzing, and testing multifunction radar systems. Reference examples provide a starting point for implementing airborne, ground-based, shipborne, and automotive radar systems. Radar Toolbox supports multiple workflows, including requirements analysis, design, deployment, and field data analysis.

AUTOSAR Blockset updates enable the use of schema version 4.4 for import and export of ARXML files and generation of AUTOSAR-compliant C code. It also offers C++-based Linux executables for adaptive models, helping to create an AUTOSAR adaptive executable to run as a standalone application.

Q Latest updates for Auto customers from MathWorks?

MathWorks recently concluded our largest event to date – MATLAB EXPO 2021. This event was executed in a virtual format with coverage over four time zones. This was the first time we brought MAC – MathWorks Automotive Conference – to the Indian audience in a friendly time zone. In case you have missed the event you can find the proceedings here – MathWorks and Frost & Sullivan recently co-hosted a well-attended Auto Executive RoundTable on ‘Challenges in simulation for automated driving development.’ You can watch the recording here – roundtable-with-industry-experts-on-challenges-in-simulation for-automated-driving-development–1613152223970. html?s_tid=srchtitle

Working along with our Accelerator partners, MathWorks launched our Technical mentoring program for startups this year. We are already working with a couple of Auto Startups in helping them accelerate their product journeys.


Niloy Banerjee

A generic movie-buff, passionate and professional with print journalism, serving editorial verticals on Technical and B2B segments, crude rover and writer on business happenings, spare time playing physical and digital forms of games; a love with philosophy is perennial as trying to archive pebbles from the ocean of literature. Lastly, a connoisseur in making and eating palatable cuisines.

Related Articles