Autonomous & Self-Driving Vehicle Vehicle News: Apex.AI, Inceptio, Baidu, Ouster, K-FLEX, indie Semi & Volvo

In autonomous and self-driving vehicle news are Apex.AI, Inceptio, Baidu, Ouster, K-FLEX, indie Semi & Volvo.

Apex.AI for Krone & Lemeken

Apex.AI, a company that develops safety-certified software for mobility and autonomous applications, announced a joint product development project for commercial-ready autonomous farming systems with Krone, a manufacturer of agricultural machinery and precision agricultural technology, and Lemken, a company that manufactures innovative machines for tillage, sowing and sustainable plant care. The ‘Combined Powers’ concept vehicle, which was developed by Krone and Lemken, is an autonomous drive unit that acts as a smart system that can plow, cultivate, sow, mow, turn and swath. The companies are now transitioning the concept carrier vehicle from the prototype stage to series production by leveraging Apex.AI’s Software Development Kit (SDK), consisting of Apex.Grace and Apex.Ida.

The commercial-ready ‘Combined Powers’ vehicle, which has already proven its performance in the fields, will save farmers valuable time in the future while maintaining the precise quality of work. There is a shortage of skilled workers in agriculture as in many other sectors. Therefore, farmers are increasingly relying on digitization and automation in order to make optimal use of the available work capacity and are increasingly leveraging innovative technology including robots that assist in livestock breeding or machines that work the fields highly autonomously.

Key software functions of the ‘Combined Powers’ architecture are based on the Robot Operating System ROS, making it a logical decision to rely on Apex.AI products, which are also based on ROS. Unlike ROS, however, Apex.AI’s products are certified by TÜV Nord for applications with special functional safety requirements and are therefore suitable for commercial vehicle use. Due to the compatibility, the data obtained with the prototypes can be used directly in series development. This makes the Apex.AI software the ideal core component for the further development of the ‘Combined Powers’ products into series production.

“Apex.AI is establishing an operating software for the autonomous age. The partnership with Krone and Lemken is a win-win for all parties: Our development environment, consisting of Apex.Grace and Apex.Ida, is the perfect basis for the applications of Krone’s and Lemken’s carrier vehicles. We are now working closely with both companies on a first project to familiarize the teams from the agricultural technology manufacturers with the suitability for series production and the possibilities of the Apex.AI products and methods. We will also evaluate our software as the framework and middleware solutions as part of the implementation phase of the VTE concept,” explains Jan Becker, CEO of Apex.AI.

“In working with Apex.AI, we see the opportunity to move from a development solution based on ROS and ROS 2 to a series solution based on Apex.Grace and being able to implement secure functions,” explains Manuel Volk, developer at Krone, the partnership with Apex.AI, “I also expect a faster learning curve in the company, since many developers are familiar with the concepts and tools of the Apex.AI SDK from their previous work with ROS and ROS 2.”

“Apex.AI enables us to develop applications that are largely independent of the hardware used. This accelerates the development process and reduces complexity,” Michael Nienhaus, a developer at Lemken, explains the collaboration.

Apex.AI, a US company with German roots, is specialized in the development of operating systems for autonomous vehicles and has extensive knowledge in the fields of robotics and artificial intelligence. Various partners in the agricultural and automotive world (e.g., AGCOMOIA, Toyota) are already benefiting from the experience. Apex.AI’s agile and iterative software development kit (SDK) enables customers to significantly increase the speed of software development and is tailored to individual applications.

Inceptio Tech Major Partners

Inceptio Technology (“Inceptio,” or the “Company”), China’s leading developer of autonomous driving technologies for heavy-duty trucks, announced new agreements with major logistics and insurance partners, shared key data points from over 50 million kilometers of accident-free autonomous driving, and showcased the core technologies that power the Inceptio Autonomous Driving System’s Truck Navigate-on-Autopilot (T-NOA) capability.

At the Company’s second annual Tech Day in Shanghai, Inceptio announced new procurement and strategic collaboration agreements with major logistics companies STO Express (SZSE:2468), ZTO Freight and Deppon Express. As part of these deals, STO Express has ordered 500 Inceptio autonomous trucks jointly developed with Dongfeng Commercial Vehicle (DFCV) and ZTO Freight has ordered 200 Inceptio autonomous trucks jointly developed with Sinotruk. Inceptio also announced a cooperation agreement with China Pacific Insurance Co., Ltd. (CPIC) that aims to develop innovative new insurance products tailored to autonomous heavy-duty trucks.

During the event, Inceptio presented the results of two new joint studies confirming the significant safety and driver experience benefits enjoyed by operators of Inceptio autonomous trucks:

  • Inceptio and CPIC jointly released the industry’s first annual insurance data safety report, which found that Inceptio’s trucks perform 75-99% better than human-operated trucks across a range of safety indicators. In particular, Inceptio trucks registered just 0.1 collision warnings per 100 kilometers, which is 98% fewer than human-operated trucks.
  • Inceptio and a team of academics published a pioneering report monitoring truck driver fatigue levels on 134 trips covering nearly 120,000 kilometers of commercial operations. The study found that Inceptio’s human safety operators experienced 35% less physiological fatigue and 11% less psychological fatigue than conventional truck drivers.

These study results demonstrate that Inceptio autonomous trucks are delivering on the four key value propositions they offer heavy-truck operators: superior safety, reduced labor costs, improved driver experiences, and better fuel efficiency. From 50 million kilometers of commercial operations, Inceptio’s partners have realized labor cost savings of 20-50% and fuel savings of 2-10%.

Inceptio autonomous trucks come equipped with the T-NOA feature, and receive regular over-the-air (OTA) updates as the Inceptio Autonomous Driving System improves itself. Inceptio’s T-NOA feature offers 100% coverage of China’s line haul network and Inceptio already has commercial business covering 70% of that network. Three core elements enable this technology:

  • End-to-end network with safety guardrails: the traditional autonomous driving software stack with discrete perception, prediction, planning, and control modules is being replaced by an end-to-end network that is both smart and reliable. Keys to this novel network are 1) guardrails to ensure the reliability and safety of network output; and 2) an efficient occupancy grid map-based representation with significantly reduced computing power and memory consumption.
  • Inceptio Super Driver: a vast trove of real-world driving data has been used to train a customized large-language model dubbed TruckGPT, allowing Inceptio’s virtual intelligent driver to surpass human drivers’ decision-making ability in a wide range of scenarios.
  • Inceptio Autonomous Truck Platform: includes a next-gen autonomous driving control unit (ADCU) designed for heavy-duty trucks and suitable for long-distance use in harsh conditions with weak wireless signals; software with unique features that significantly enhance development efficiency and can be adapted to new vehicle models in just 9-12 months; and truck electrical and electronic architecture (EEA) with new features including full modularity with decoupling of software and hardware, facilitating efficient upgrades.

Inceptio founder and CEO Julian Ma said: “After another year of hard work and momentous achievements, we couldn’t be more excited to share Inceptio’s progress with the world. We are truly proud of the great strides we have made to commercialize our technology, making nearly 50,000 trips on 340 routes for more than 100 freight and logistics customers.

The new orders we announced represent a huge vote of confidence from our valued partners STO Express, ZTO Freight, and Deppon Express, which have all experienced the benefits of our technology first-hand. And through our new alliance with CPIC, we are developing insurance solutions that will help accelerate the mass adoption of autonomous trucks even further. We look forward to delivering more mass-produced L3 autonomous trucks to our partners in the near future as we continue striving to make freight transport greener, safer, and more reliable.”

New Research in Autonomous Driving with Florida Universities

Four years after announcing an innovative partnership on autonomous vehicle (AV) technology, Florida Polytechnic University and Tallin University of Technology (TalTech) in Estonia have taken their research to the next level. The collaboration has produced significant results, enabling the exchange of knowledge, resources, and expertise between the two institutions and beyond.

The partnership integrates Florida Poly’s expertise in AV validation and verification with TalTech’s ability to run AV technology on a shuttle built on open-source software, which allows access to the software’s internal workings. This created an open-source environment called PolyVerif, which provides new tools for testing and validating AV technology.

“There’s a need for a research platform where we can accelerate the rate at which we’re solving the safety problems with AV technology, and that’s what our joint research with TalTech offers,” said Dr. Rahul Razdan, senior director for special projects at Florida Poly and researcher at its Advanced Mobility Institute (AMI).

Razdan said the research has been used by the Jacksonville Transportation Authority, with whom AMI also has an ongoing partnership. On the TalTech side, the initial work led to the formation of the commercial entity Auvetech, which offers AV shuttles worldwide.

“It’s been rewarding to see the progress we’ve been making through our partnership with Florida Poly’s Advanced Mobility Institute in developing tools that are helping in the advancement of AV technology,” said Raivo Sell, a robotics professor who leads the AV research group at TalTech. “We’re looking forward to continuing working together on solving critical challenges in safety, testing, and verification.”

As part of the collaboration, Sell spent months at Florida Poly working with faculty and building a robust research framework. The work provided crucial information that facilitated a National Science Foundation grant awarded to Florida Poly’s AMI.

Florida Poly and TalTech have garnered attention beyond their partnership. TalTech submitted a proposal to the Baltic American Freedom Foundation to sponsor several talks by Razdan as an expert in the field. These were hosted in EstoniaFinland, and Latvia, including one at FinEst Centre for Smart Cities, a multi-national research organization focused on improving urban environments by testing and developing new technologies.

Baidu Expands Apollo Go

Baidu, Inc. (NASDAQ: BIDU and HKEX: 9888) (“Baidu” or the “Company”), a leading AI company with strong internet foundation, announced the expansion of Apollo Go, its autonomous ride-hailing platform, to expand its driverless car service to cover Wuhan Tianhe International Airport. This marks the first time in China that an autonomous ride-hailing service has been established between urban areas and an airport, as well as the first instance of Chinese autonomous vehicles connecting both urban roads and highways. The service is currently offered to selected Apollo Go users by invitation, and will be made available to the general public in September.

The expansion of Apollo Go’s operation area to Wuhan Tianhe Airport is a significant step towards pioneering driverless airport transportation in China. It also means Baidu has now unlocked more challenging operation scenarios for its robotaxi fleet, making autonomous ride hailing service more accessible to the public.

The significant number of passengers traveling through Wuhan’s Tianhe Airport provides a strong foundation for the further growth of Baidu Apollo. Located 25 kilometers from the center of Wuhan, Tianhe Airport is one of China’s eight major regional hub airports. This year, Wuhan Tianhe International Airport has transported a total of 125,000 international regional passengers, placing it first in the central China region. The airport’s daily flight limit has increased from 700 to approximately 1000 flights, making it the leader in flight scheduling capacity in the central China region and one of the top airports nationwide.

Following the expansion to cover the airport area, Apollo Go will continue to expand its presence in Wuhan, with plans to bring its services to the city’s Jingkai District, Hanyang District, East and West Lake District, Qiaokou District, and other core areas of the Jiangbei area in the future.

As a leading autonomous ride-hailing service provider, Apollo Go has recorded over 3.3 million cumulative orders as of June 30, 2023. In the second quarter of this year alone, Apollo Go provided 714,000 rides, a 149% YoY increase. Today, Baidu and Apollo Go are advocating for the widespread application of fully autonomous vehicles. Baidu’s fully driverless robotaxi fleet is now operating in five cities, including BeijingShenzhen, and Wuhan. It is the first company to conduct completely fully autonomous driving commercial operations and testing in multiple cities throughout the country.

Ouster Problems with Hesai

Ouster, Inc. (NYSE: OUST) (“Ouster” or the “Company”), a leading provider of high-performance lidar sensors, issued the following statement today relating to the Company’s complaint filed with the U.S International Trade Commission (“ITC”) and the investigation into the unfair trade practices of Hesai Group (Nasdaq: HSAI) and related entities (ITC Investigation):

The presiding Administrative Law Judge (ALJ) recommended, following a motion by Hesai, that the investigation be terminated to allow arbitrators time to decide whether Ouster is required to arbitrate based on a prior Settlement Agreement between Velodyne and Hesai entered in 2020, before Ouster and Velodyne merged in February 2023. The initial determination is not a decision on the merits of Ouster’s ongoing patent infringement case against Hesai.

The motion is the latest attempt by Hesai to avoid a ruling regarding whether its imported lidar products infringe Ouster’s intellectual property rights and part of a larger pattern of delay, including invoking the Chinese Data Security Law during discovery.

Ouster invented digital lidar technology following an engineering breakthrough and holds one of the largest patent families in the lidar industry. Ouster’s complaint sets forth how, after the market shifted toward Ouster’s digital lidar, Hesai stole Ouster’s revolutionary patented technologies and incorporated them into Hesai’s competing products.

Ouster previously filed a patent infringement action in the United States District Court for the District of Delaware. That case is stayed pending the ITC investigation. Should the ITC investigation be terminated, the mandatory stay in the District Court of Delaware will be lifted and the patent infringement case will commence.

Ouster welcomes that the investigation will now be reviewed by the ITC Commissioners. The company will continue to vigorously enforce its patents and seeks to bar all infringing products from the United States.

K-FLEX Autonomous Reliability Results

How can autonomous driving be made more reliable? On August 30, the “KI-FLEX” project, which was funded by the German Federal Ministry of Education and Research (BMBF) and led by Fraunhofer IIS, presented its research results. The initiative was built around a high-performance, energy-efficient, and yet flexible hardware platform with the corresponding software framework, which uses AI technology to process and fuse data from various sensors. This allows vehicles to perceive and localize environmental stimuli in a manner that is fast, efficient, and reliable.

If autonomous vehicles are to make the correct decision in every conceivable situation, they must be able not only to locate their own position in traffic, but also to reliably capture their environment with precision. To do this, vehicles must have the ability to collect and fuse data from sources such as laser, camera, and radar sensors. For the algorithms that process such sensor data, artificial neural networks have become indispensable tools. However, these networks need fast, efficient, flexible hardware, which is precisely what the “KI-FLEX” project has been successfully researching over the past four years. “This is an important step toward the safe mobility of the future,” says Michael Rothe, who heads the Embedded AI group at Fraunhofer IIS.

Systems must be able to unambiguously detect and identify objects and road users in traffic situations. As such, the importance and utility of the individual sensors vary accordingly. Both the traffic situation and the weather and light conditions have to be taken into account in order to ensure safe autonomous driving. Moreover, the systems must be able to respond flexibly to potential sensor failures or adversarial attacks in their data. To address this requirement, the project partners developed resource-optimized approaches for the early and late fusion of camera data, lidar data, and detected objects along with an AI-based monitoring system. These components allow vehicles to reliably respond to changed situations by adjusting the algorithms used.

Artificial neural networks are currently developing at a rapid rate. The growing number of architectures is making increasing demands on the hardware and software. For this reason, “KI-FLEX” employs a heterogeneous hardware architecture made up of FPGA and ASIC AI accelerators in order to implement the neural networks for object detection in camera and lidar data. This flexibly reconfigurable and programmable AI accelerator system anticipates the future to some extent, as the hardware will be able to support emerging neural network designs. Furthermore, the hardware platform’s computing resources can be allocated dynamically according to load.

The AI chip developed in the project also offers considerable advantages with regard to power consumption, processing speed, and cost savings compared to conventional multi-purpose processors (CPUs) or graphics processing units (GPUs).

The project “KI-FLEX – Reconfigurable hardware platform for AI-based sensor data processing for autonomous driving,” which launched in September 2019, was funded by the German Federal Ministry of Education and Research (BMBF) within the guidelines on promoting research initiatives in the field of “AI-based electronic solutions for safe autonomous driving (AI element: autonomous driving).”

Led by Fraunhofer IIS, the project consortium comprises several German research and industry partners: Infineon Technologies AG, videantis GmbH, the Technical University of Munich (Chair of Robotics, Artificial Intelligence and Real-Time Systems), the Fraunhofer Institute for Open Communication Systems FOKUS, the Daimler Center for Automotive IT Innovations (DCAITI, Technical University of Berlin) and the Friedrich-Alexander-Universität Erlangen-Nürnberg (Chair of Computer Science 3: Computer Architecture).

indie Semi RFE Silicon Transceiver

indie Semiconductor (Nasdaq: INDI), an Autotech solutions innovator, has launched the world’s first commercial fully integrated 240 GHz radar front-end (RFE) silicon transceiver. This device builds upon the success of indie’s previously released 120 GHz solution, expanding the Company’s portfolio of short-range, high precision and high resolution terahertz frequency radar solutions.

As a complement to the well deployed use of 76 GHz to 81 GHz radar for long range automotive sensing, recent safety initiatives such as European New Car Assessment Program (Euro NCAP) are driving the use of 120 GHz terahertz frequency radar for in-cabin driver and occupant monitoring due to its higher resolution. The higher frequency-enabled superior precision of 240 GHz radar is being leveraged for new and rapidly emerging vehicle dynamics and monitoring applications, including assessment and control of air spring-based suspension settings, real time road surface quality and hazard assessment to dynamically adapt ride quality, and even fine grade monitoring of gas tank levels.

In addition to the growing automotive market opportunity, the resolution and bandwidth benefits of 240 GHz radar are also applicable to industrial sensing applications including material thickness measurement and analysis, end of line product quality, tank level monitoring (now accelerated through recent EU regulation), surface inspection and security scanners. indie’s 240 GHz solution can readily address these industrial terahertz frequency applications, which Mordor Intelligence estimates will be a $1.8 billion electronics opportunity by 2028.

“The launch of our 240 GHz transceiver solution marks a significant leap forward in radar technology,” Dr. Peter Gulden, SVP of indie Semiconductor’s Radar Systems and Software. “We are proud to offer a solution that not only enhances safety within vehicle cabins but also revolutionizes industrial monitoring applications. Our latest high frequency radar is a game-changer in resolution, integration and system cost.”

The TRA240091 is a cascadable radar front-end with an operating bandwidth of up to 45 GHz at 240 GHz. These unique features enable extremely high-resolution which is ideal for applications within the license free 244 – 246 GHz ISM band and beyond.

Volvo Autonomous Goes Driverless

Volvo Autonomous Solutions (V.A.S.) achieves industry-first milestone with the removal of the safety driver in an active commercial mining operation at Brönnöy Kalk mine in Velfjord, Norway.

The autonomous transport solution developed for Brönnöy Kalk consists of seven fully autonomous Volvo FH trucks and V.A.S.’s in-house developed virtual driver. Operating in challenging conditions that include steep inclination, extreme weather and long stretches of dark tunnels, the trucks haul limestone from the mine to the crusher.

The Brönnöy Kalk project at a glance

  • The solution includes seven Volvo FH Trucks, V.A.S. in-house developed virtual driver for confined areas, infrastructure, training as well as a comprehensive repair and maintenance program.
  • The trucks are used to transport limestone between the mine and crusher on a five kilometer stretch that covers tunnels and outdoor environment. The wheel loader operator uses a touch screen in the wheel loader to call the trucks for loading and to manage the operation.