Driving Electric Vehicle Affordability: How AI and Machine Vision Lower Costs
Electric vehicles (EVs) are here to stay. More manufacturers are committed to reducing vehicle emissions, going carbon neutral, and introducing more electric vehicles into their lineups.
Consumer demand for EVs is growing too, driven by government incentives such as tax breaks, investments in domestic manufacturing, and setting goals for more new EV sales. But the road to mass EV adoption is long; less than 1% of the 250 million vehicles in the U.S. are electric, according to Reuters, and only 17 million new cars are sold annually.
Sticker Shock Stalling Growth
One of the biggest consumer speed bumps is the price tag. EVs can have lower maintenance and fuel costs than their gasoline-powered counterparts, but the upfront cost of an EV is still driving away motorists, despite tax rebates such as those included in the Inflation Reduction Act and other incentives.
While the average transaction price (ATP) of EVs is declining, they’re still more expensive than vehicles with internal combustion engines (ICE). Late last year, Kelly Blue Book estimated the ATP for new EVs at $61,448, about 34% higher than non-luxury ICE vehicles at $45,578.
That trend could continue into 2030, according to Ford Motor Co. CEO Jim Farley. At an investor conference in May 2023, Farley said EV prices are not expected to decline until second and third-generation models are released later in the decade.
Race to the Bottom: Lower Battery Prices, Greater Market Share
The high costs of EVs are inhibiting widescale acceptance and ultimately limiting the number of vehicles and batteries on the market. As countries incentivize EV purchases and manufacturing competition accelerates, EV producers and suppliers will be motivated to reduce costs and pass lower prices onto consumers. Tesla, for example, reduced prices on specific models by as much as $2,800 in May, according to Investor’s Business Daily, aiming to capture more market share.
“It's pretty straightforward," Tesla CEO Elon Musk said at an annual meeting in May. "We see what the demand is, and then we adjust pricing to meet the demand."
Two significant cost centers for EVs are batteries and labor. Lithium-ion battery packs, which can account for up to 50% of an EV's price tag, according to Reuters, require expensive, rare materials such as cobalt, nickel, and lithium. Prices for those metals are volatile and affected by policy changes from major EV and metal-producing countries, among other factors, Forbes reports.
Financial services firm Morningstar anticipates lithium prices to remain elevated until 2030 as EV demand grows and new lithium producers, like any large-scale operation, generally face delays.
The sentiment is more clear-cut when it comes to labor costs. Ford’s Farley anticipates EVs will become easier to build and outfitted with smaller batteries using more cost-effective materials, lowering labor costs between 2030 and 2035.
Steering Into Cost Savings
EV manufacturers and battery suppliers are compelled to drive down costs. Some of the most attainable ways to lower costs are by automating complex, time-consuming EV battery inspections and manual processes.
Machine vision and AI are crucial for automation in manufacturing, enhancing efficiency, quality control, and cost reduction. Machine vision utilizes cameras and sensors to capture and analyze visual data, while AI interprets this data to solve complex and challenging inspections.
Differentiating Between Acceptable and Authentic Defects
These technologies lower EV battery costs by reducing inspection times, lowering scrap rates, and identifying subtle defects to safeguard battery quality and efficiency.
Identifying flaws is essential for EV battery safety and performance. Defect detection is an essential function, but many processes are time-consuming, resource-intensive, and unreliable. Classifying a cosmetic blemish as a functional flaw is a costly mistake; it can mean scrapping expensive, rare metals and elements.
To minimize waste and rework, Cognex utilizes advanced algorithms and image analysis software to separate genuine defects from superficial flaws during EV battery production. Using deep learning technology, users can program a vision system to detect defects, determine if a flaw is within the acceptable range of variation, and flag unacceptable defects while accounting for variations such as reflective surfaces.
Assessing Battery Welds
Inspecting EV battery weld seams is critical for EV structural integrity and performance. It’s almost impossible to separate cosmetics from functional variations with a traditional vision system because their appearances are nearly identical.
Cognex software locates the inspection area, while 3D sensors inspect welded edges and corners to ensure they are entirely sealed and defect-free. Deep-learning based defect detection and classification tools are trained on a wide range of weld variations and learn to accurately classify and distinguish functional flaws from cosmetic ones.
Inspecting Battery Cell, Pouch, and Cylinder Surfaces
A thorough surface inspection is vital to eliminating defects, contaminants, and anomalies that could impact battery performance. Machine vision systems detect imperfections like scratches, dents, or foreign particles, improving the overall quality of EV batteries.
Inspecting Electrode Coatings
The uniformity and quality of electrode coatings significantly influences EV battery performance. Precise machine vision solutions inspect electrode coatings, identifying inconsistencies or defects. This ensures uniform coating thickness and quality, enhancing battery performance and longevity.
Cognex industrial line scan cameras are ideal for “web” surface inspections to ensure that the film substrate is coated evenly with copper and aluminum.
Machine Vision Ready to Contribute to EV Affordability
Leveraging machine vision and AI solutions is crucial for EV manufacturers to address challenges and improve efficiency and production quality. By adopting these technologies, manufacturers can reduce costs, increase productivity, and accelerate EV adoption.
For detailed information on implementing machine vision and AI in EV manufacturing, download the EV Solutions Guide below.