In autonomous and self-driving vehicle news are NHTSA, Tesla, STRADVISION, Valeo, Mobileye and Here.
NHSTA Tesla AutoPilot Recall
Tesla is recalling over two million vehicles in the U.S. for an update to the Autopilot system. The system is supposed to make sure drivers are paying attention. It was found to be defective. Tesla will send out a software update to fix the issue. The recall follows a two-year investigation by NHTSA into crashes during the Autopilot system was in use, some of which were fatal. Drivers are misusing the system on roads the wrong kinds of roads. The recall includes models Y, S, 3, and X made between October 5, 2012, and December 7, 2023.
Gatik Supplies Autonomous Box Trucks for Kroger in TX
In a significant step forward for autonomous B2B logistics, Gatik AI Inc. (Gatik) and The Kroger Co. (NYSE: KR) announced the successful launch of the companies’ multi-year commercial collaboration, which sees Gatik’s medium-duty autonomous box trucks transporting fresh, customer-favorite products from Kroger’s state-of-the-art Customer Fulfillment Center (CFC) in Dallas, Texas, to multiple retail locations.
To celebrate the remarkable strides made since beginning local operations earlier this year, the companies came together for a ribbon-cutting ceremony with City of Dallas and Texas state officials including Tennell Atkins, Mayor Pro Tem and Council Member, and representatives from Texas Department of Transportation and Texas Department of Public Safety.
Over the past eight months, Gatik has safely and consistently made delivery runs multiple times per day, seven days per week across Kroger’s Dallas distribution network. The operations have increased the speed, responsiveness, and frequency of fulfilling Kroger e-commerce orders, further solidifying Kroger’s commitment to providing a seamless shopping experience for its customers.
The state of Texas has emerged as a national leader in the deployment of autonomous technology, and the support from municipal and state government representatives has been instrumental in facilitating Gatik’s successful operations with Kroger. This support underscores the state’s commitment to fostering technological innovation, advancing the development of a high-performance transportation ecosystem and ensuring unparalleled economic growth across the state’s supply chain.
Gatik’s continued growth in the Texas market underscores its dedication to advancing the regional supply chain, providing safe and efficient autonomous middle mile delivery services to several Fortune 500 companies in the state. In 2024 through 2025, the company is on track to achieve Freight-Only operations at scale, which involves deploying Gatik’s autonomous trucks without human operators on board, ensuring the true benefits of autonomous transportation are realized for local communities.
STRADVISION Study Advances Research
STRADVISION, an automotive industry pioneer in deep learning-based vision perception technology, announced that its CTO, Jack Sim, presented a groundbreaking study, “Efficient Deep Learning Development through Teacher-Student Learning and Sprints and Its Application to 3D Perception Systems,” at Tech.AD USA 2023. The event, focused on ‘level 3 to X autonomy & autonomous vehicle technologies,’ took place from December 6 to 8, 2023, at The Henry, Dearborn, Michigan.
In his presentation, Sim unveiled an efficient software development process for multiple deep-learning tasks, leveraging teacher-student learning and sprints. This methodology, concurrently improving accuracy and speed, resulted in a remarkable reduction of product development time by up to 50%.
The process was successfully applied in the creation of STRADVISION’s augmented reality product, ImmersiView™. Sim explained that the approach decouples dependencies between tasks, trains individual teacher networks in sprints for improved accuracy, and employs a unified student network for simultaneous task learning.
“To optimize accuracy while maintaining inference speed constraints, STRADVISION applied neural architecture search to the student network in the initial sprints and subsequently fixed the architecture in the remaining sprints,” Sim explained. This efficient process enabled the first prototype creation within three months, followed by consistent monthly software releases.
The methodology was further extended to address the challenges of building multi-modal 3D perception systems, demonstrating its adaptability and effectiveness in handling diverse sensor inputs during training. Sim emphasized, “This approach ensures comprehensive and accurate perception, exemplifying the adaptability and effectiveness of our development process in handling diverse sensor inputs in training.”
Jack Sim concluded, “Our research underscores the significance of our method in advancing the field of multi-modal perception in autonomous systems, paving the way for the future of automotive technology.”
STRADVISION continues to lead the way in deep learning-based vision perception technology, driving innovation and setting new standards for efficient and adaptive development processes.
Valeo & Mobileye Combine Tech
Valeo, world leader in advanced driving assistance systems (ADAS), started its collaboration with Mobileye in 2015, choosing to integrate into its front camera system hardware and software the Mobileye system-on-chip (SoC), named “EyeQ®”. Mobileye, with partners like Valeo, revolutionized driving assistance systems by taking computer vision technology to a next level in the automotive industry. Together, Valeo and Mobileye have combined their best in class technology and have developed and manufactured multiple generations of front camera systems. The strong collaboration between Valeo and Mobileye is now focused on the integration of the latest generation of the Mobileye SoC into the Valeo front camera system and also into Valeo’s cutting edge centralized computer and software.
The Valeo’s front camera system using the Mobileye EyeQ contributes to the safety of road users by supporting key features such as autonomous emergency braking, adaptive cruise control, lane keeping assist, and traffic sign recognition. A front camera system is the key enabler in reaching safety requirements defined by authorities around the world, and we expect that by 2030 about 90% of all new cars will be equipped with advanced driver assistance technology.
Marc Vrecko, President of Valeo’s Comfort and Driving Assistance Systems Business Group, said: “Valeo offers the largest portfolio of sensors and perception software available on the market. Front cameras are one of the key technologies for enhanced safety and will soon be set up on all vehicles. We are very proud to have reached this new milestone in the ramp up of our production and in our partnership with Mobileye. At Valeo, we strive to make our technology efficient, but also available to all and we are proud to be contributing to the safety of road users“.
“Since our founding we have focused on improving vehicle safety for drivers globally. Our ongoing success with Valeo demonstrates how far we’ve come together and the opportunies that lie ahead to bring even more advanced technologies to millions of vehicles over the next few years. We look forward to working with Valeo with the next-generation Mobileye EyeQ6 system-on-chip going into full production next year“, said Nimrod Nehushtan, Senior Vice President of Strategy and Business Development at Mobileye.
Today, more than 90% of road accidents are caused by human error. ADAS are at the heart of the transformation of mobility, driven by enhancements in safety and comfort. The ADAS market is expected to grow by 17% per year to reach 60 billion euros in 2030. Valeo has the market’s most comprehensive portfolio of sensors (ultrasonic sensors, cameras, radars and LiDARs), software and associated intelligence. Beyond their collaboration for cameras, Mobileye and Valeo announced in September a new collaboration on software-defined imaging radars to support the next-generation of driver assist and automated driving features.
Here Supplies HD Live Map for BMW 7 for Level 3
Here Technologies, the leading location data and technology platform, today announced the usage of the Here HD Live Map onboard BMW’s new 7 Series with “BMW Personal Pilot” Level 3 highly automated driving function. This marks a key milestone in the companies’ longstanding collaboration on high-definition mapping for advanced driving assistance and automated vehicle systems.
BMW’s Personal Pilot allows drivers to take their hands off the wheel and eyes off the road under certain conditions. Drivers can use the extra time to work, chat or even use entertainment features such as video streaming while the function is activated. Beginning with the new 7 Series, BMW will introduce the function starting in Germany in March 2024.
The Here HD Live Map provides safety-enhancing data to the vehicle system for tasks such as localization, prediction and path planning, working in combination with various vehicle sensors. The Here HD Live Map extends the functions capabilities to “see” and “plan” in scenarios where vehicle sensors alone might fall short, enhancing overall situational awareness, by delivering precise information about the road network’s features and local rules in advance.
BMW also utilizes Here HD map data to define the operational design domain (ODD), which lets the vehicle know when and where the Personal Pilot function can be activated.
“The Here HD Live Map plays a central role for our BMW Personal Pilot Level 3 functionality. It acts as the primary input source to create the car’s driving path, providing critical information about attributes such as geometry, lane positioning and predicting road traffic signs in time”, said Nicolai Martin, SVP Driving Experience at BMW Group. “With the BMW Personal Pilot, the car drives itself ‘on-map’, with the car’s sensors acting as an extra layer sharing the safety load with the map.”
“We are proud to support our partner BMW in bringing the next level of vehicle automation to customers,” said Jason Jameson, Executive Vice President and Chief Customer Officer at Here Technologies. “The integration of the Here HD Live Map with BMW’s Personal Pilot function marks a significant achievement, underlining the pivotal role high-definition maps play in the future of automated driving.”
The Here HD Live Map contains three distinct layers:
- Road Model: provides detailed road characteristics, such as topology, direction of travel, elevation, slope, road rules, boundaries, tunnels and intersections.
- HD Lane Model: contains precise lane-level information, such as locations of lane lines, widths, markings, boundaries, access characteristics, stop areas, raised surfaces, and speed limits.
- Localization layer: classifies roadside furniture, such as signs, barriers, poles, signals, road surface marking, places of interest and overhead structures.
The Here HD Live Map is based on UniMap, Here’s next-generation map-making capability, which features the unique ability to rapidly fuse large quantities of data from diverse sources, including vehicle cameras, industrial lidar and satellites, into a fully aligned and unified global map.
The BMW Personal Pilot is one of the first commercially available Level 3 highly automated driving functions. Starting in March 2024 it will first be available in the new BMW 7 Series to be used on motorways in Germany. When activated, the system will allow a vehicle to drive itself in conditionally automated mode at a maximum speed of 60km/h in heavy traffic.
Here is the global market leader for automotive-grade maps designed to power navigation, advanced driving assistance (ADAS), EVs and automated driving systems. Today, approximately 180 million cars have been sold with Here maps on board. This includes more than 40 million cars across Europe and North America that use Here location data and services for ADAS and automated driving functions.