3D + AI in EV Battery Inspection

Chris Aden, Sr. Director of AI Solutions, LMI Technologies

The growing demand for EV batteries has already permanently transformed the car manufacturing landscape, with almost all major automotive companies now operating electric vehicle production lines. While electric vehicles do have less moving parts than their internal combustion engine (ICE) counterparts, manufacturing them is still extremely complex. From stators and rotors, to intricate wiring harnesses, LMI is involved with many EV manufacturing applications. But the heart of an EV is its battery, and this is where machine vision solutions have become essential to market success. In this presentation an LMI Technologies speaker will look into the different ways 3D scanning and inspection is used in EV battery manufacturing to ensure maximum product quality, longevity, and safety. The discussion will include 3D sensing applications in key EV manufacturing steps such as (1) electrode manufacturing, (2) cell assembly and packaging, (3) cell-to-module assembly, (4) module-to-pack assembly, and (5) final installation inspection. We will cover how 3D smart sensors and their built in, onboard measurement tools are used in every stage of these manufacturing processes. The discussion will center around the ability of 3D imaging to generate critical shape data (i.e., height information) of the target object that is not possible with 2D inspection that can only capture intensity-based information (i.e. greyscale). In addition, we will discuss the ability to combine LMI's AI deep learning-driven service with its proven 3D sensor technology to produce an even more accurate, reliable, scaleable inspection solution. We will outline how LMI's AI service provides a turnkey solution to help you leverage the power of AI in your factory. We tackle data collection, pipeline training and development, user interface design, and factory communication. Once deployed, we continue to maintain and optimize the system to ensure your return on investment is fully realized. Attendees will learn how EV Battery quality inspection is particularly suitable for AI due to its repetitive nature and high level of predictability. This makes it possible to collect a dataset of images that can be used to train a custom neural network for classification, object, or anomaly detection. These datasets can be trained and retrained to continuously optimize for their inspection task over time. The presentation will showcase how products that are nearly identical and need to be measured for tolerance or conformity are probably easier to solve using a stack of traditional algorithms, whereas products such as EV batteries that require a subjective evaluation, sometimes from an experienced visual inspector, are highly suitable for deep learning based inspection systems.  Attendees will come away with a deeper understanding of how 3D (for image acquisition) and AI deep learning sy

Back