Proceedings

Latha Chembrakalam, Continental Automotive India

Moderator: Zahida Vaseem, MathWorks

India has a thriving startup system and many of these startups have gone on to become leading players in the Indian industry ecosystem. Some start-ups are incubated in “academic incubators.” Statistics show that over 50% of the incubators in the country are associated with educational institutions. Join this panel discussion to hear from incubation center experts, industry players, and academicians to understand:

• why and how academia can contribute to the needs of industry and startups in emerging technology trends
• how innovation in engineering pedagogy can have an impact on the talent pool
• how engineers and scientists of tomorrow can be prepared to address grand challenges
• the need for new modes of collaboration between industry and academia

Moderator: Prashant Rao, MathWorks

AI is gradually becoming mainstream with development teams looking at it by design rather than as an afterthought. The use of AI in industries including automobile, aerospace, manufacturing, and medical devices is advancing engineering and research and accelerating product development. Embedding AI into machines and systems makes them intelligent. AI is now broadly distributed and exists on the device, on the edge, and on the cloud. This gives enterprises significant flexibility and the opportunity to develop transformative applications that predict the remaining useful life of machines, perform remote surgeries, develop automated driver assistance systems, and more. However, the development and deployment of AI in these embedded devices is not without challenges such as limited skills, non-availability of data and data complexity, the lack of tools to build models, and integration of AI projects to scale. This panel explores the adoption of AI technologies in multiple industries and how organizations are tackling these challenges.

Dr. Rishu Gupta, MathWorks
Peeyush Pankaj, MathWorks

Drifting data poses three problems: detecting and assessing drift-related model performance degradation; generating a more accurate model from the new data; and deploying a new model into an existing machine-learning pipeline. See a solution that addresses each of these challenges through an example of a real-world predictive maintenance problem. We reduce the complexity and costs of operating the system—as well as increase its reliability—by automating both drift detection and data labeling. Learn how to develop streaming analytics on a desktop, deploy those solutions to the cloud, and apply AutoML strategies to keep your models up-to-date and their predictions as accurate as possible.

Highlights:

• Developing an end-to-end MLOps pipeline
• Monitoring an AI model from drifting data/ behaviour
• Generating automatic training datafrom incoming data<
• Retraining and redeploying models
• Use case: Battery SOC estimation

Uvaraj Natarajan, MathWorks
Jayanth Balaji Avanashilingam, MathWorks

Wireless communication is continuously evolving to make revolutionary changes across various applications and industries. New innovative approaches like artificial intelligence (AI) will solve traditional wireless problems and help achieve higher performance and efficiency. In this session, see how easy it is to apply AI capabilities to solve wireless communications problems in MATLAB^®. You will learn how to be more efficient by leveraging various AI techniques, ready-to-use algorithms, and data generated with MATLAB.

Shripad Chandrachood, MathWorks
Vikrant Singh, MathWorks

Battery-powered vehicles are rapidly emerging as the future of mobility with the continued advances in manufacturing and charging technologies. As the engineers innovate to decarbonize energy production and transform mobility, simulation studies can play an important role in accelerating this transition.

In this master class, take a deep dive in battery and EV charger modeling, its integration with the electric grid, and perform system level simulation studies. The examples highlight the Model-Based Design approach, which can be a powerful enabler to test what-if scenarios and bring in efficiency in development. Explore the following topics:

• Battery System Design and Development:
  • Building detailed battery cell and pack models along with a cooling system
  • Performing thermal analysis and system level simulation of a BEV/FCEV
  • Developing a battery management system (BMS)
• EV Charging:
  • AC/DC charging techniques and associated power electronics control design
  • Developing a ‘Fast DC Charger’ model using Simscape Electrical™
• Microgrid Simulation:
  • Simulating microgrids with renewables (solar/wind), EVs, and energy storage systems
  • Utilizing controller hardware and real-time simulation to test and validate energy management algorithms

Nukul Sehgal, MathWorks
Rajat Arora, MathWorks

The automotive industry is accelerating its investments in electrification, autonomous driving, connected vehicles, and modern user experience. These trends demand more computing power and innovative electric, electronic, and complex software architectures in cars.

The automotive industry is embracing service-oriented architectures (SOA) as a new paradigm to design software applications that are highly reusable, easy to update, and loosely coupled to hardware. SOAs are based on the concept that an application consists of a set of services that are dynamically discovered, published, subscribed, and reconfigured at run time. SOA concepts are used in multiple industry standards, including AUTOSAR, ROS, and DDS.

Highlights:

• What is SOA?
• Modeling and simulation of SOA-based services in Simulink™
• Adaptive AUTOSAR in Simulink