Large Language Models (LLMs) have shown significant potential in automating code generation tasks offering new opportunities across software engineering domains. However, their practical application remains limited due to hallucinations - outputs that appear plausible but are factually incorrect, unverifiable or nonsensical. This paper investigates hallucination phenomena in the context of code generation with a specific focus on the automotive domain. A case study is presented that evaluates multiple code LLMs for three different prompting complexities ranging from a minimal one-liner prompt to a prompt with Covesa Vehicle Signal Specifications (VSS) as additional context and finally to a prompt with an additional code skeleton. The evaluation reveals a high frequency of syntax violations, invalid reference errors and API knowledge conflicts in state-of-the-art models GPT-4.1, Codex and GPT-4o. Among the evaluated models, only GPT-4.1 and GPT-4o were able to produce a correct solution when given the most context-rich prompt. Simpler prompting strategies failed to yield a working result, even after multiple refinement iterations. These findings highlight the need for effective mitigation techniques to ensure the safe and reliable use of LLM generated code, especially in safety-critical domains such as automotive software systems.

This is the shocking moment a Tesla in'Full Self-Driving' (FSD) mode plowed into a deer standing in the middle of the road. The driver, Paul S, did not confirm when or where the crash occurred, or what model Tesla he was driving. But dashcam footage shows the vehicle driving down a clear two-lane highway at night moments before the animal suddenly came into view. The Tesla rammed directly into the deer, without stopping or slowing down'even after hitting the deer on full speed,' Paul said. 'Huge surprise after getting a dozen false stops every day!' he added.

Federated learning (FL) has emerged as a promising approach for training machine learning models on decentralized data without compromising data privacy. In this paper, we propose a FL algorithm for object detection in quality inspection tasks using YOLOv5 as the object detection algorithm and Federated Averaging (FedAvg) as the FL algorithm. We apply this approach to a manufacturing use-case where multiple factories/clients contribute data for training a global object detection model while preserving data privacy on a non-IID dataset. Our experiments demonstrate that our FL approach achieves better generalization performance on the overall clients' test dataset and generates improved bounding boxes around the objects compared to models trained using local clients' datasets. This work showcases the potential of FL for quality inspection tasks in the manufacturing industry and provides valuable insights into the performance and feasibility of utilizing YOLOv5 and FedAvg for federated object detection.

Federated learning (FL) has gained significant traction as a privacy-preserving algorithm, but the underlying resemblances of federated learning algorithms like Federated averaging (FedAvg) or Federated SGD (Fed SGD) to ensemble learning algorithms have not been fully explored. The purpose of this paper is to examine the application of FL to object detection as a method to enhance generalizability, and to compare its performance against a centralized training approach for an object detection algorithm. Specifically, we investigate the performance of a YOLOv5 model trained using FL across multiple clients and employ a random sampling strategy without replacement, so each client holds a portion of the same dataset used for centralized training. Our experimental results showcase the superior efficiency of the FL object detector's global model in generating accurate bounding boxes for unseen objects, with the test set being a mixture of objects from two distinct clients not represented in the training dataset. These findings suggest that FL can be viewed from an ensemble algorithm perspective, akin to a synergistic blend of Bagging and Boosting techniques. As a result, FL can be seen not only as a method to enhance privacy, but also as a method to enhance the performance of a machine learning model.

With unique styling and a motorized, rotatable camera the 70mai Dash Cam Omni both looks good in your vehicle and takes detailed captures in any direction. I don't want to saddle the 70mai Dash Cam Omni with the adjective cute, though the onboard and app graphics animations push that aesthetic somewhat. Clever and unique are far better terms to describe the unit with its motorized, rotatable camera. More than that, the Omni is an effective dash cam that takes detailed video from any direction you point it at--day or night. Note that all images are showing the Omni resting its mount.

The $300,000 Cadillac Celestiq is the brand's bid to reclaim the "Standard of the World" title, and it will be equipped with what General Motors thinks will be the best driver assistance technology. The sleek, electric four-door will be the first GM product to feature Ultra Cruise, which is a step above the Super Cruise system that's available today and is being developed to provide hands-off driving on most roads 95% of the time. Super Cruise offers hands-off driving on 400,000 miles of pre-certified highways using radars, cameras, GPS and hyper-accurate maps while facial recognition tech ensures the driver is paying attention and ready to take control when required. Jason Ditman, Ultra Cruise chief engineer, said that Ultra Cruise will have a forward-looking lidar unit mounted behind the windshield that work along with both short-range and long-range radars, long range cameras and over 20 sensors in total to provide full coverage of what's around the car. The Celestiq's lidar will be installed behind the windshield.

Worldwide, the most popular car colors are white, black and gray, in that order--underlying just how boring auto buyers are. Part of this is likely a desire not to stand out, and part of it is undoubtedly that sober-colored cars perform best when it comes to long-term value retention. BMW is hoping to fix this chromatic conundrum with a new concept car that not only changes color on command, but also uses e-ink to make facial expressions with its grille and headlamps. It can even project a head-up display across the entire windshield. Shown off at the CES technology show in Las Vegas, the BMW i Vision Dee is also surprisingly compact.

Tesla is recalling nearly 27,000 cars in the US due to windshield defrosting problems, according to a US safety regulator. The electric vehicle company is recalling 26,681 cars, including some 2021-2022 Model 3, Model S, Model X, and 2020-2022 Model Y vehicles, according to a safety recall report. In an acknowledgement letter from the US National Highway Traffic Safety Administration (NHTSA) dated 8 February, Tesla informed the federal organisation that a software error linked to the vehicle's heat pump was behind the windshield defrosting. "A software error may cause a valve in the heat pump to open unintentionally and trap the refrigerant inside the evaporator, resulting in decreased defrosting performance," the letter said. The defrosting problem may reduce drivers' visibility and potentially increase the risk of a crash, the NHTSA noted.

Automotive head-up displays (HUDs), systems that transparently project critical vehicle information into the driver's field of vision, were developed originally for military aviation use, with the origin of the name stemming from a pilot being able to view information with his or her head positioned "up" and looking forward, rather than positioned "down" to look at the cockpit gauges and instruments. The HUD projects and superimposes data in the pilot's natural field of view (FOV), providing the added benefit of eliminating the pilot's need to refocus when switching between the outside view and the instruments, which can impact reaction time, efficiency, and safety, particularly in combat situations. In cars, the main concern is distracted driving, or the act of taking the driver's attention away from the road. According to the National Highway Transportation Safety Administration, distracted driving claimed 3,142 lives in 2019, the most recent year for which statistics have been published. Looking away from the road for even five seconds at a speed of 55 mph is the equivalent of driving the length of a football field with one's eyes closed.

Nuro's delivery autonomous vehicles (AVs) don't have a human driver on board. The company's founders Dave Ferguson (president) and Jiajun Zhu's (CEO) vision of a driverless delivery vehicle sought to do away with a lot of the stuff that is essential for a normal car to have, like doors and airbags and even a steering wheel. They built an AV that spared no room in the narrow chassis for a driver's seat, and had no need for an accelerator, windshield or brake pedals. So when the company petitioned the U.S. government in 2018 for a minor exemption from rules requiring a rearview mirror, backup camera and a windshield, Nuro might have assumed the process wouldn't be very arduous. In a 2019 letter to the U.S. Department of Transportation, The American Association of Motor Vehicle Administrators (AAMVA) "[wondered] about the description of pedestrian'crumple zones,' and whether this may impact the vehicle's crash-worthiness in the event of a vehicle-to-vehicle crash. Even in the absence of passengers, AAMVA has concerns about cargo ejection from the vehicle and how Nuro envisions protections from loose loads affecting the driving public."