Controls Specialist - Engine Engineering, Controls Engineering

Ford set to lose ground to Hyundai in Q3 Sales, says Cox Automotive - 'Detroit 3' market share seen hitting record low. |Remove ads. Asian brands are forecast to capture more than half of U. S. new-vehicle sales, Cox said. General Motors is expected to remain the top-selling U. S. automaker in Q3, despite declining sales. Cox expects Tesla's U. S. sales in Q3 to slump more than 30%. Detroit's traditional auto giants are expected to lose further ground to Asian rivals in the third quarter, with Cox Automotive forecasting Ford (F) to fall behind Hyundai Motor Group (HYMTF) in U. Discover more S. vehicle sales. At the time of writing, F shares were down 1. 2%, tracking a fifth straight session of decline, while GM dropped more than 2%. Cox Sees Q3 Sales Decline For Ford, GM And TeslaCox expects Ford to sell 504,172 vehicles in the third quarter (Q3), down 7. 1% from last year, while Hyundai is forecast to sell 511,421 vehicles, up 6. 5% year over year. General Motors (GM) is expected to remain the top-selling U. S. automaker in Q3, despite sales expected to fall more than 5. 2%. Ford's weakness follows a soft second quarter in which U. S. sales fell 10%, partly due to phasing out vehicle models. Cox expects Ford's year-to-date sales through September to fall 8. 8% while GM's sales are forecast to decline 6. 2%. It also expects Tesla's (TSLA) U. S. Q3 sales to slump more than 30%. Hybrid Vehicle Demand Boost Asian CarmakersCox sees Asian automakers continuing to gain ground as U. S. consumers increasingly shift toward hybrid vehicles and passenger cars, segments where Asian brands have traditionally held a stronger position. Asian automakers are expected to account for more than half of U. S. new-vehicle sales for a second straight quarter. Together, the 'Detroit 3,' which also includes Stellantis (STLA), are expected to hold just over 36% of the market in Q3, their lowest share on record. Ford Looks Beyond CarsFord, however, has a new trick up its sleeve - expansion into the battery storage market through Ford Energy. The company has committed roughly $2 billion to transform its Kentucky battery operations to manufacture energy-storage systems. Ford has also signed a five-year agreement allowing EDF Power Solutions to procure up to 20 GWh of battery storage systems, with deliveries expected from 2028. Sean Duffy Urges Ford To Reduce China Dependency Ford has also faced criticism from the Trump administration over its ties with Chinese companies. Earlier this month, Transportation Secretary Sean Duffy urged Ford to reduce its reliance on China, raising concerns about its use of CATL battery technology in Michigan, its partnership with Geely in Spain, and its continued production of some Lincoln vehicles in China. Ford responded by saying its CATL arrangement is a technology-licensing agreement rather than a joint venture, and that it owns and operates the Michigan battery plant with American workers. Morgan Stanley said that while Duffy's criticism creates a new regulatory overhang, it is "likely more benign than feared. "Retail's Take On Ford And GMRetail sentiment surrounding F has remained 'neutral' over the past 24 hours, while sentiment for GM trended in the 'bearish' territory. F shares are down 3. 9% so far in 2026, while GM stock has edged 1. 1% higher. Both stocks have outperformed Tesla, which is down 13% year-to-date. Full story hosted on AutoWorld - Updated from official sources Enjoying this story? Solve the math problem to verify you're human. By submitting you consent to receive communications from AutoWorld. By subscribing you consent to receive the AutoWorld Daily Brief. Ready to find your next vehicle? Get personalized quotes or browse listings from sellers nationwide. Rate this story

Hybrid vehicle-to-grid algorithms fail to beat simpler alternatives, study finds. Peer-reviewed study finds hybrid V2G scheduling algorithms offer no advantage over simpler APSO approach, questioning vendor claims in smart charging market. September 24, 2026 V2G is currently being explored across the globe, including Australia, with the likes of Ausgrid trialing the technology. A peer-reviewed study testing seven optimisation algorithms for scheduling EV charging and discharging in vehicle-to-grid (V2G) systems has found that combining two algorithms into a "hybrid" does not automatically produce better results than using the stronger of the two on its own. The study, published in the journal Processes, was conducted by researchers at the University of Engineering & Technology Lahore, the University of Management and Technology Lahore, the University of Johannesburg and the University of Botswana. The researchers tested the algorithms against a simulated smart parking facility where EVs arrive and depart across a scheduling window, each with its own battery capacity and driver-preferred charge level, under a time-of-use electricity pricing scheme with peak, mid-peak and off-peak periods. Particle Swarm Optimization (PSO) and the Whale Optimization Algorithm (WOA) are both metaheuristic algorithms, meaning they search for a near-optimal solution to a complex problem by mimicking a natural process rather than solving it exactly. PSO models a swarm of candidate solutions that move through the search space influenced by their own best-known position and the swarm's collective best position, while WOA is modelled on the bubble-net hunting behaviour of humpback whales, alternating between encircling and spiralling movements to close in on a solution. Adaptive Particle Swarm Optimization (APSO) is a refined version of PSO that continuously adjusts its own internal search behaviour, tightening or loosening its search pattern in response to how clustered or spread out its candidate solutions are at each step, rather than following a fixed search pattern throughout. APSO was the best-performing algorithm in the study, improving on standard PSO by 8% and on a PSO-WOA hybrid, which combines elements of both algorithms in an attempt to capture the strengths of each, by 9.7%, while also producing a measurably flatter peak load demand profile across the simulated charging facility. The researchers found the PSO-WOA hybrid performed no better than plain PSO once tested statistically, and a separately enhanced version of WOA could not be shown to outperform the standard WOA either. Of the seven algorithms tested, only APSO's advantage held up consistently after applying the full battery of statistical tests across the 100 simulation trials for each algorithm. The researchers ran parametric tests, including t-tests and ANOVA, and non-parametric tests, including Mann-Whitney U, Wilcoxon Signed-Rank, and Friedman tests, and post-hoc corrections, including Holm's Step-Down, Bonferroni-Dunn, and Nemenyi analyses, specifically to distinguish genuine performance improvements from results that could plausibly have occurred by chance. The paper argues this level of validation is uncommon in existing V2G scheduling literature, noting that most prior studies report performance improvements without statistical testing to confirm that those improvements are real. The researchers conclude that, for this class of scheduling problem, the effective mechanism behind APSO's advantage is its adaptive control of internal search parameters, which adjust based on the distribution of the algorithm's search population at each iteration, rather than the act of hybridising two algorithms. That distinction is directly relevant to vendors marketing "hybrid" or multi-algorithm optimisation as inherently superior to a single, well-tuned approach. Vendor performance claims contrast with independently validated trials. The findings arrive as V2G and smart charging platforms increasingly lean on algorithmic and AI-branded capability as a point of commercial differentiation. As reported by EV Infrastructure News, Geely recently launched an AI-powered fast-charging system, Xingrui PowerMind, built with Chinese AI company StepFun, which the company says predicts battery temperature up to 30 seconds ahead to dynamically adjust charging power. Geely claims a 20% improvement in battery lifecycle when the system is combined with its pulse-restoration technology. As with much of the smart charging software market, those figures come from the manufacturer's own testing rather than independent statistical validation of the kind applied in the Processes study. The commercial stakes for getting scheduling right are rising as more automakers commit to V2G on a global scale. Hyundai Motor Group and energy intelligence platform Kaluza launched a partnership in August 2026 to embed smart charging into the Kia and Hyundai apps, starting in the UK and Australia, with V2G services planned to follow from 2027. Kaluza CEO Stephen Fitzpatrick described the partnership as laying the technical foundation for how EVs integrate with the energy system more broadly. The real-world value of that claim will depend on how well the underlying scheduling software performs against driver charging preferences and grid price signals under live conditions, precisely the kind of question the Processes paper argues needs independent statistical testing rather than vendor-reported figures alone. By contrast, one of the clearer examples of rigorously validated smart charging in the market to date comes not from a vendor claim but from a randomised controlled trial. The Centre for Net Zero, an Octopus Energy-founded research institute, ran a 12-month RCT across more than 13,000 UK households and found that AI-managed charging cut peak household electricity use by 42% and reduced annual bills by £650 (US$887), with more than half of participating households never manually overriding the automated system. That trial's design, a control group tested against four treatment groups and tracked over a full year using high-frequency smart meter data, offers a template for the kind of evidence base the Processes researchers argue is largely missing from the algorithmic V2G literature more broadly. Not every claim in the smart charging market has been put through this kind of scrutiny. The Processes study is itself not without limitations. Battery degradation costs and charging-efficiency losses are explicitly excluded from its model, and the entire parking facility is treated as a single connection point to the grid rather than modelling feeder- or transformer-level constraints. The researchers acknowledge that incorporating these factors would likely yield more conservative cost estimates and flag the work as a direction for future model refinement rather than a claim that the study reflects a full commercial deployment. In that sense, the paper's core contribution is methodological rather than a ready-made blueprint for production scheduling software. It demonstrates how to rigorously test competing algorithms, more than it proves that any single algorithm is ready to be deployed at scale. Senior Reporter George joined Solar Media in August 2022, writing for its UK sites, Solar Power Portal, and EV Infrastructure News'. After a spell as Editor for the UK sites, George relocated to Sydney, Australia, to support its APAC expansion. Get the latest EV infrastructure updates direct to your inbox Read expert news, data-driven analysis and incisive opinions to help you navigate tomorrow's EV infrastructure landscape. Google Preferred Source