The stack is primarily based on deep learning AI processing for all of its modular components, including environmental perception, sensor fusion and vehicle path planning. It was designed in conjunction with the company’s CV3-AD AI domain controller SoC to ensure it runs optimally on the SoC’s CVflow AI engines – minimizing power consumption and the processing load for its on-chip Arm cores. The SoCs also provide optimal processing for vision and radar sensing and fusion, while Ambarella’s stack more broadly offers a flexible implementation model, enabling OEMs to leverage any or all of its modules in combination with their own software.

A key feature of the AI-based software stack highlighted by Ambarella is its deep-learning-based planner, a dynamic vehicle path planning module that enables a more natural driving experience. In developing and refining this functionality, Ambarella leveraged its evolving R&D vehicle fleet which allowed the deep planner module to be trained with data based on a wide variety of real-world scenarios. The latest vehicles in this fleet are equipped with a sensing suite consisting of mono and stereo cameras, as well as Ambarella’s Oculii 4D imaging radar, with the CV3-AD processing the data sent from these sensors. While highlighting this configuration, Ambarella noted that its solutions are flexible to support OEM’s specific sensing suites (including the option for adding LiDAR).

Another feature of the stack highlighted by Ambarella was its requirement of standard-definition maps, eliminating the need for pre-generated HD maps. Here, HD maps are generated in real time using live environmental data fed from the sensing suite. This capability offers support in handling difficult AD scenarios, such as downtown areas in large cities with roundabouts, narrow roads with parked vehicles, heavy traffic, construction, and a high density of pedestrians, cyclists and other vulnerable road users.

The CV3-AD SoCs run Ambarella’s entire AD stack with enhanced power consumption, AI performance per watt, and simplified thermal management. The company said that, because its software and hardware are designed to work optimally together, OEMs can sufficiently integrate additional software to create differentiated features, including those for functional safety.

Together, the companies will collaborate on enhanced high-resolution radar systems for automotive applications by leveraging Distributed Aperture Radar (DAR), a technology developed by Zendar that enables a simplified system solution approach. Through DAR, the resolution of high-performance radar systems can be enhanced while eliminating the need for thousands of antenna channels. It collects and combines information from multiple radar sensors on a vehicle to create a larger effective antenna that achieves heightened sensing resolution. DAR also enables high-angular resolution below 0.5 degrees, providing capabilities similar to a LiDAR that support precise mapping of the environment, and offers further potential to enhance resolution through flexible mounting techniques.

The jointly-developed radar solutions will be based on NXP’s S32R radar processor platform and RFCMOS SAF8x one-chip SoCs – both of which are tailored to cover a variety of vehicle architectures and help accelerate the transition toward distributed architectures. The companies expect that OEMs and suppliers will equally benefit from high resolution systems developed with simplified standard radars, with reduced thermal complexity, and a smaller radar footprint for flexible and more convenient mounting.

Priced between RMB 3,000 ($411) to RMB 3,999 ($547), the HP170 enables autonomous driving on highways and expressways, automatic emergency braking, and short-distance memory parking. These functions are enabled by its sensor system, consisting of 12 ultrasonic radars, four fisheye cameras, two rear corner radars (as optional), one forward facing camera, and a forward-facing radar as well as 5 TOPS of computing power.

The HP370, priced between RMB 5,000 ($684) to RMB 5,999 ($821), offers 32 TOPS of computing power, two forward facing cameras, two side-view cameras, one rear-view camera, four fisheye cameras, one front radar, two rear corner radars, 12 ultrasonic radars and two front corner radars as optional. It supports additional functions, such as memory driving within a certain city, and comes with intelligent barrier avoidance.

At the top of the range is the HP570, priced between RMB 8,000 ($1,095) and RMB 8,999 ($1,232), which offers customers the choice between a chip capable of 72 TOPS, and one capable of 100 TOPS. Its sensor suite consists of two forward facing cameras, four side-view cameras, one rear-view camera, four fisheye cameras, one front radar, 12 ultrasonic radars, and optionally allows for LiDAR sensor to be added. The HP570 enables automatic driving on urban roads and is equipped with an all-scenario parking assist.

It combines various automotive sensors, such as HDR cameras and 3D LiDARs, and industrial-grade computers into one package that offers a suite of tools for processing sensor fusion, easing and accelerating the development of perception solutions. In addition to Tier IV’s key systems, the Kit also includes documentation detailing its integration and operation.

In developing the Kit, Tier IV is likewise aiming for users to avoid complex undertakings related to sensor fusion – such as sensor and computer configurations, ROS2 environmental setups, system synchronization and calibration processes, and perception model tunings. Avoiding these can help reduce development costs and expedite time-to-market. The software provided with the Kit is open source, allowing users to employ it as a base for their own projects.

Among the key capabilities of the Kit include its compatibility with Autoware, the open-source project for autonomous driving and its configurability – allowing for the customization of solutions with Tier IV’s automotive HDR cameras, ECUs, LiDARs, and more to suit various project requirements. The company’s open-source tools, enabled and powered by Tier IV’s sensor calibration and synchronization capabilities, work to minimize development times, while its ready-to-deploy hardware and autonomous applications accelerate product development.

In the future, Tier IV plans to expand its Kit to support a wider range of sensors and computers, updating the machine learning models and algorithms for the corresponding perception technology to cater to a greater range of use cases.

The data fidelity and range upgrades offered in 4Sight+ likewise support enhancements in dependability, performance, and adaptability that work to help OEMs eliminate phantom braking, improve existing ADAS features, and safely enable hands-free driving functions at highway speeds.

4Sight+ takes advantage of the platform’s modular, bistatic, system that can quickly reconfigure LiDAR performance and add new capabilities over-the-air. The sensor itself is powered by a small MEMS device that facilitates ultra-fast movements, while the larger 4Sight Intelligent Sensing Platform is designed to be highly programmable and meets the performance requirements for both urban and highway driving.

The sensor likewise improves prediction capabilities by tracking vulnerable road users – including pedestrians, cyclists, and other vehicles – at a distance of up to 300 meters. It improves vehicle ride quality and reaction times by detecting small objects, such as bricks and tires, at up to 200 meters, while supporting vehicle path planning by identifying road surfaces, including cement and asphalt, at up to 100 meters, even in direct sunlight and low light environments. These advantages will help AEye enable predictive safety systems for smoother, more comfortable, hands-free driving experiences operating at more than 80 mph (128 km/h).

In leveraging a software-defined sensor, AEye will look to benefit from using the same hardware to handle a broad range of initiatives and objectives – including close-up vehicle cut-ins, long-range views, and more. Through 4Sight+, AEye is aiming for OEMs to leverage its scanning agility to quickly reconfigure LiDAR performance – using cues from other sensors in the vehicle, such as speed or GPS location, to activate uniquely defined modes of operation for parking, urban, and highway environments.

By optimizing LiDAR performance based on the driving conditions, the sensor operates efficiently, with a high performance at the lowest power required of it. This further allows for flexibility in form factors, a quality that will support seamless vehicle integration for OEMs pursuing electrification.

The OX02C1S is a 2.5-megapixel (MP) RGB-IR BSI global shutter (GS) sensor for in-cabin driver and occupant monitoring systems (DMS, OMS). The OX01H1B is a 1.5MP mono (IR) GS sensor for driver monitoring systems (DMS). With a pixel size of just 2.2 microns (µm), both offer enhanced NIR quantum efficiency (QE) at 36%, representing an increase in modulation transfer function over the previous 3.0 µm FSI GS pixel design, and low power consumption to support high-performance capabilities.

Both the OX02C1S and OX01H1B are compact, automotive-grade, sensors that leverage their enhanced QE to support low-light performance. Both, likewise, utilize OmniPixel’s 4-GS technology – which enables simultaneous image detection in all pixels to accurately reproduce rapid motion without any deformation. They also have integrated ASIL-B and cybersecurity that meet the latest industry standards. Both come in an OMNIVISION stacked a-CSP package, allowing for higher-performance image sensors in tighter camera spaces, and are available in a reconstructed wafer option for designers that are looking for the flexibility to customize their own package.

The OX02C1S sensor features 2.5MP resolution in a 1/3.55-inch optical format, while the OX01H1B features 1.5MP resolution in a 1/4.51-inch optical format. OmniVision confirmed that samples of its new GS sensors are available now, and are expected to enter mass production in 2024.

CARIAD selected the InnovizTwo next-generation LiDAR sensor and perception software.

This is Innoviz’s third design win with a major automaker and first deal as a Tier-1 supplier. With this win, Innoviz’s forward-looking order book was updated to $6.6 billion

The company selected Fabrinet based on its deep expertise in advanced optical packaging and precision optical. This in addition to its knowledge of electro-mechanical and electronic manufacturing services, through which its LiDARs will be produced and assembled.

The partnership represents a strategic move for Cepton, accelerating its goal to make LiDAR an essential component put to use in everyday consumer vehicles. Prior to the announcement, the company had begun initial production of Vista-X90 assemblies at Fabrinet’s facility in Chonburi, Thailand in September 2021. The facility is automotive qualified with IATF 16949 certification.

The Vista-X90 itself features Cepton’s patented MMT (Micro Motion Technology), a LiDAR imaging mechanism that enables long-range, high-resolution, 3D imaging through a dense scan pattern. Free of rotation, mirror, and friction, MMT was designed to provide the industry with scalable LiDAR solutions that carefully balance performance, reliability and cost.

Vista-X90 also packages Cepton’s proprietary LiDAR engine ASICs, designed for optimal illumination control, detection and signal processing. In utilizing Fabrinet’s experience in high-volume optical and opto-electrical manufacturing, the company aims to deliver high-quality, cost-effective, assemblies to ramp up production and achieve scalable growth.

Outside, the doors leverage a smart ISD interactive system and UWB wireless communication technology with 10cm-level positioning. This allows the EV to automatically detect people, keys, and surrounding vehicles.

Inside the cabin features a fully digital setup headed by HiPhi Bot, an AI companion capable of automatically adjusting many aspects of the user experience. The assistant interacts with passengers and comes with a 8-direction, adjustable, robotic arm that can perform a variety of delicate movements . Further along the interior is a 23-speaker Meridian audio system and racing bucket seats.

Accelerating from 0 to 100 km/h (62 mph) in 3.8 seconds, the car comes with rear-wheel steering and an IVC vehicle dynamic control system that generates a steering angle of 13.2 degrees. The vehicle’s 120kWh high-performance battery pack leverages Cell to Pack technology and cooling liquids to offer heat diffusion and help prevent fires while providing up to 705 km of battery life.

Powering the vehicle’s ADAS is the OEM’s proprietary HiPhi Pilot system, which provides dual-redundancy for perception, communication, braking, steering, and power supply. In further enabling advanced safety safety, the HiPhi Z leverages an NVIDIA Orin X chip and BlackBerry’s QNX Neutrino OS to utilize 34 sensors and LiDAR technology to constantly scan its surroundings and perform assisted driving and parking functions. This combination of technologies also enables the vehicle’s smart summon feature, which see the vehicle autonomously drive to the user’s location.

The HiPhi Z’s pricing will range from $89,000 to $129,000 and, in the lead-up to further details being announced, it will be on display in Shenzhen from July 6. Human Horizons plans to debut the EV further across 11 key major cities. The OEM also confirmed that the HiPhi Z’s launch date will be announced at the Chengdu Auto Show in August, where interested users will be able to test drive the vehicle.

Valeo’s third-generation LiDAR sees everything, even when an object is far ahead, invisible to the human eye. It can identify objects more than 150 meters ahead that neither the human eye, cameras nor radars can, such as a small object with very low reflectivity, for example a tire lying in the road. It recreates a 3D image of the vehicle’s surroundings using a point cloud, with an as yet unparalleled resolution for an automotive system. It can map the ground topology and detect road markings.

Valeo LiDAR also features embedded high performance software based on artificial intelligence algorithms, which enables it to steer the vehicle’s trajectory, anticipating obstacle-free zones on the road ahead. It can self-diagnose and triggers its cleaning system when its field of vision is obstructed. Like all Valeo LiDAR, the technology is automotive grade, meaning that the data it generates remain fully reliable and accurate in all usage and weather conditions (from -40 up to +85°C). It stands as the key component in a sensor system enabling vehicles to achieve approval for SAE conditional driving automation (level 3), meeting the legal requirements of UN-R157.

Valeo’s third generation LiDAR makes driving safer and gives time back to the driver in bothersome driving situations, such as when traveling at low or medium speed in heavy traffic. These challenges are central to the partnership between Stellantis and Valeo. Through its data collection capabilities, this LiDAR will enable new services to be offered to Stellantis’ customers.

Valeo is already world number one in advanced driver assistance systems (ADAS), equipping one in three new cars globally with its technologies. It was the first, and to date remains the only, company to produce an automotive LiDAR scanner on an industrial scale. More than 170,000 units have been produced and the technology is protected by more than 500 patents. The Group intends to accelerate even further in this field, as announced in February 2022 with the launch of its Move Up plan, the value creation strategy at the heart of the four megatrends disrupting mobility (electrification, ADAS, reinvention of the interior experience and lighting).

Upon receiving the permit, Vueron conducted a test in which its LiDAR drove a vehicle autonomously from Los Angeles to San Francisco. The test vehicle equipped with the company’s LiDAR completed a 383-mile route that ran through Interstate 5 and Interstate 580, driving for six hours at a maximum speed of 70 mph.

Vueron has said that the safety driver inside Vueron’s vehicle did not hold the steering wheel for the duration of the test, and that no further autonomous assistance was provided by other technologies, such as cameras, radars, GPS, or high-definition mapping systems. Likewise, Vueron advised that it further verified the solution through self-driving tests conducted in a variety of roadway conditions and environments.

In Q3 2022, Vueron expects to unveil the next generation of its LiDAR platforms that can be worked on immediately by its customers. The company is also aiming to provide these systems at an affordable price point without compromising on delivering a safe, secure, driving experience.

The first camera rolled off the Irish production lines in 2002. Eight years later, in 2010, one million units were delivered to customers. In 2013, the ten-millionth camera threshold was hit, and in 2015, the figure doubled to 20 million. Between 2015 and 2018, output increased 30-fold compared with the first eight years, with 30 million cameras manufactured in three years, for a total of 50 million. The four years between 2018 and 2022 saw total production double and cross the 100 million camera mark. The Tuam site is at the heart of this industrial success, to which three other Valeo plants around the world have contributed (another in Europe, one in North America, one in Asia).

These 100 million cameras would not have been possible if Valeo had not successfully accomplished three revolutions. The first was in the field of industrial performance, where mass production made high-tech systems widely accessible across the world. Then came the safety and comfort revolution, which made it possible to see what was previously invisible and thereby ensure safer and easier mobility. Last was the intelligence revolution, when artificial intelligence algorithms and the associated software became core components of Valeo’s vision systems, enabling new capabilities such as through-the-body vision and autonomous parking and driving features.

Valeo has an extensive Research and Development center integrated into the Tuam site with a wide range of Product and Process Engineering skills in Mechatronics, Systems & Software as well as Test and Validation. More than 70% of its staff are engineers with coding expertise who have developed innovative features for the automotive industry. One example is image fusion technology, which creates a single snapshot from multiple video sources. This is how today, when drivers in many cars around the world shift into reverse, they can see their vehicle as if it were filmed from above by a drone – even though no such drone exists. The stream of images filmed, recorded and processed by artificial intelligence provides enough data for cars to reach level 2+ automation. In 2022, an important worldwide Valeo customer will release a vehicle equipped with this smart camera. It will have advanced assistance and automation features, both for parking (low-speed maneuvers) and for driving.

Valeo is already number one in the world in advanced driving assistance systems (ADAS), equipping one in three new cars globally with its technologies, and intends to accelerate even further in this field. Its cameras play an essential role in this acceleration. The Group announced this objective in February 2022 with the launch of its Move Up plan, the value creation strategy at the heart of the four megatrends that are disrupting mobility (electrification, ADAS, reinvention of the interior experience and lighting).

As early as in 2019, Hesai’s Pandar series lidars were integrated onto WeRide’s first batch of hundred-unit Robotaxi fleets, successfully entering road testing and operation stage. Hesai’s lidar products have also supported WeRide to commercialize China’s first Robotaxi service to the public in Huangpu District, Guangzhou. The Robotaxi fleet has taxi operation permits and charges the same rate as any normal taxi in Guangzhou. The commercial operation has entered its third year of service. Hesai and WeRide’s collaboration keeps evolving. Hesai’s lidar sensors have been integrated into WeRide’s new autonomous driving products including Mini Robobus, Robovan, and Robosweeper.

As of January 2022, WeRide has reached over 10 million kilometers of autonomous driving on public roads, combining testing and operations with more than 300 AD vehicles. WeRide’s fleets comprise Robotaxi, Robobus, Robovan, and Robosweeper, running in cities such as Guangzhou, Shenzhen, Shanghai, Zhengzhou, Nanjing, Wuhan, Boao in China, San Francisco in the US, and Abu Dhabi in UAE, being tested and operated daily. 

The automotive grade, hybrid solid-state lidar AT128 uses Hesai’s self-proprietary chip technology, integrating 128 laser channels to achieve 1200×128 overall resolution and  200-meter detection range (at 10% reflectivity). It can effectively capture details of the objects within FOV to provide powerful perception capabilities for high level autonomous driving, further accelerating the iteration of autonomous driving algorithms. At the same time, the highly integrated transmit-receiver module and the stable scanning structure makes AT128 one of the most competitive and comprehensive lidars on the market. AT128 will be mass produced at Hesai’s new manufacturing center Maxwell, which will commence operation in 2022 with a planned capacity of over 1 million units per year.

Nissan’s “ground truth perception” technology fuses information from next-generation high-performance LIDAR, radar and cameras. The technology can detect the shape and distance of objects, as well as the structure of the area surrounding the vehicle, in real time with a high degree of accuracy. Utilizing this information, it is possible for the vehicle to instantly analyze the current situation, judge and automatically perform required collision-avoidance operations. This technology can also detect slowed traffic and road obstacles in the distance and execute lane changes accordingly. Importantly, the technology can also provide increased support to drivers in areas where detailed map information is not available.

Nissan has partnered with cutting-edge companies to research and develop this technology and share know-how. The next-generation LIDAR, a crucial element, is being integrated into Nissan’s system in collaboration with Luminar, utilizing world-leading technologies. In the area of advanced collision avoidance control it is critical to develop highly accurate verification technology in a digital environment. To this end Nissan is teaming up with Applied Intuition, which has state-of-the-art simulation technology.

Under its long-term vision, Nissan Ambition 2030, Nissan is working to develop vehicle control technology aimed to significantly reduce accidents by utilizing next-generation LIDAR technology. Nissan believes that in the upcoming era of autonomous driving, driver-assistance technology that can avoid highly complex accidents will be essential. Nissan aims to complete the development of its ground truth perception technology by the mid-2020s. It will first be available on select new models, and on virtually every new model by fiscal year 2030.

Various lighting situations on the road may pose obstacles to the driver. Quick transitions from a low-lit environment to a brighter one, such as exiting a tunnel, may require a few seconds for the driver’s eyes to adjust. Also, flickering from LED headlamps or road signage, such as street lamps and traffic lights, could become increasingly noticeable on the automotive system’s camera screen.

The ISOCELL Auto 4AC offers a safer driving experience with an enhanced field of view for the driver with its CornerPixel technology. The technology features a specialized pixel structure that mitigates LED light over 90-hertz (Hz). Within a single pixel area, it embeds two photodiodes, one 3.0µm pixel for viewing low light images, and a 1.0µm pixel placed at the corner of the big pixel for brighter environments. With two photodiodes capturing images in different exposures simultaneously, the sensor offers up to 120dB HDR with minimal motion blur, allowing smoother transitions between dark and bright areas while preserving more details of the road ahead.

To minimize LED flickering, the smaller photodiode’s exposure time can be extended, preventing pulsing LED light from being displayed as flickering on the camera screen. This delivers a more pleasant viewing experience for the driver and more accurate image data on LED-embedded objects for the automotive system to recognize.

The Samsung ISOCELL Auto 4AC comes in a 1/3.7-inch optical format with 1.2 million 3.0-micrometer (µm) pixels, and for streamlined client system installations, an image signal processor (ISP) is embedded within the sensor.

The new test demonstrates the sensor’s capabilities in adverse weather conditions, and behind a first surface (in this case, a windshield). It was conducted to study the impact of adaptive LiDAR on the performance and safety of automated functionality. AEye’s team used a rain machine to simulate wet weather and mounted the sensor behind a piece of windshield glass to gauge long-range sensor performance in heavy rain. In a second test, its LiDAR sensor was able to detect small objects in a tunnel (through rain and a second surface) at 120 meters. It also detected five bricks and a black dog not visible to the human eye at the same distance, alongside a pedestrian and a child at 110 meters.

In March, the firm announced that VSI Labs had verified its LiDAR’s range, resolution, and speed capabilities, and its ability to place the sensor behind first surfaces (such as the windshield or grill) with a minimal impact to performance. The latter is critical to automotive OEMs, as it provides flexibility in implementing sensors within their vehicle designs without compromising on aesthetics or changing aerodynamics.

AEye’s bistatic architecture streamlines vehicle integration by separating the transmit and receive paths. This provides an optical isolation that ensures any light reflected does not blind the sensor, ensuring optimal performance – even in adverse weather conditions. This performance is further enhanced by AEye’s use of 1550 nanometer lasers that provides detection in areas with lower visibility – such as rain, snow, and smoke.

AEye’s intelligent LiDAR utilizes adaptive sensing to deliver this enhanced performance. The technology addresses the most difficult challenges facing autonomous driving while meeting functional safety requirements. AEye’s adaptive LiDAR is also capable of scanning the scene around it while focusing on important objects to enable safer, quicker, decision-making in complex scenarios.

In addition to performance testing from VSI Labs, the sensors were also validated by NTS, who put them through extreme automotive shock and vibration tests.

The solid-state Velabit sensor now has a simultaneously achievable maximum horizontal FoV of 90° and maximum vertical FOV of 70°, approximately three times more points per second than the previous model. The next generation Velabit takes all of Velodyne’s learnings to offer a configurable and dynamic field of view and delivers a high-resolution zoom capability, all within a compact and lightweight sensor.

Versatile Performance

This lightweight and versatile Velabit sensor features a small form factor and low power consumption to maximize battery life and vehicle range. The small size and low cost allow the sensor the unique ability to be utilized in multiple industries and for critical safety applications, such as:

• Automotive and new mobility
• Autonomous mobile robots
• Industrial forklift and warehouse
• Infrastructure and smart city
• Sidewalk and last-mile delivery
• UAV mapping and navigation

Engineered to be an optimal automotive-grade lidar solution for Advanced Driver Assistance Systems (ADAS) and autonomous vehicles, Velabit can fill sensing gaps, helping automakers achieve full coverage around a vehicle. The sensor enables robust perception coverage for ADAS features including Blind-Spot Monitoring, Cross-Traffic Detection, and Pedestrian Automatic Emergency Braking (PAEB). Additionally, the wide vertical field of view makes this an ideal technology for near-field sensing and robotics applications.

Software + Hardware Integration

Velabit will be combined with Velodyne’s lidar-based perception software, Vella, to streamline customers’ projects and advance a broad spectrum of vehicle safety functions. The software translates Velabit’s high-quality point cloud data into valuable perception outputs, including object classification and tracking, obstacle detection, scene segmentation, and object velocity, that can be readily utilized by vehicle systems. When combined with vehicle odometry data, Vella provides time and distance to collision measurements, enabling safer vehicle response.

The sensor can also be combined with other Velodyne lidar technologies, such as the Velarray H800, for high-speed operation or function as a standalone lidar solution in low-speed applications.

Designed for Mass Production

In combination with MLA, Velodyne’s fully automated manufacturing process and global manufacturing partnerships enable cost optimization and high-quality mass production, as early as Q4 2022.

Since its launch, Velabit has received widespread industry acclaim, including being named a winner in the 2020 Best of What’s New awards by Popular Science. The sensor also was named the winner of the 2020 Innovation Award by Silicon Valley Robotics.

Key Velodyne customers and other top global innovators in automotive and robotics technology are scheduled to attend Velodyne’s upcoming Velabit Demo Days to witness the Velabit in action, and will have an opportunity to reserve future capacity.

In combination with a suite of other sensor modalities, thermal cameras help create a safe autonomous vehicle system for mixed traffic environments, where other vehicles, bicyclists, and pedestrians are sharing the same roadway.

ADASTEC has selected FLIR thermal cameras to be included on the recently debuted flowride.ai automation platform. In combination with a suite of other sensor modalities, thermal cameras help create a safe autonomous vehicle system for mixed traffic environments, where other vehicles, bicyclists, and pedestrians are sharing the same roadway.

In conjunction with the development of autonomous passenger vehicles, OEMs and innovators are also working on public transport vehicles as part of the larger push to develop an autonomous vehicle ecosystem. One company making strides in the public transportation autonomous space is ADASTEC and it recently debuted flowride.ai, the world’s first level-four, full-size, full-speed autonomous bus platform.

The flowride.ai automation platform is designed to operate 24 hours a day, every day of the year, including through ever-changing weather conditions, requiring the need for thermal cameras from Teledyne FLIR. In combination with a suite of other sensor modalities, thermal cameras help create a safe autonomous vehicle system for mixed traffic environments, where other vehicles, bicyclists, and pedestrians are sharing the same roadway.

ADASTEC integrated thermal cameras with the primary focus on improving detection and safety of all vulnerable road users on or near the road and at bus stops. Furthermore, thermal cameras are a key sensor technology to more readily distinguish and classify pedestrians who are partially obscured by nearby vehicles and other common roadway and urban objects.

The funding will be used to increase the production capabilities of its automotive-grade LiDARs, support supply-chain partners for mass production capabilities, and further expand Innovusion’s research and development efforts around advanced LiDAR Technologies.

In January 2021, NIO announced ET7, its first vehicle with full-stack autonomous driving capabilities, scheduled to be delivered in the first quarter of 2022. The vehicle is equipped with an image-grade ultra-long-range LiDAR developed jointly by Innovusion and NIO, a key component of ET7’s NIO Autonomous Driving (NAD) system. A model application in the field of autonomous driving, the LiDAR has a 120-degree horizontal viewing angle, ultra-high resolution, and the industry’s longest detection range, capable of “seeing” up to 500 meters. Innovusion’s unique “flexible gaze” function generates a more detailed three-dimensional point-of-view to better track vehicles and pedestrians while simultaneously maintaining high-resolution across the entire field of view, even in low-light conditions.

The shuttles are expected to achieve level 4 autonomy, enable fully automated driving and be used to transport both passengers and cargo in various geo-fenced settings. The agreement is expected to generate hundreds of millions in revenue throughout the life of the program.

The new design win followed a stringent two-and-a-half-year due diligence process conducted by the Tier-1 automotive supplier. As part of the selection process, the supplier completed a thorough review of Innoviz’s product reliability, manufacturing processes and maturity, automotive qualifications, corporate processes certifications and more.

Unlike costly mechanical spinners installed on car roofs, InnovizOne is a low-cost, automotive-grade, lightweight solution. InnovizOne is designed to ensure that the shuttles circumvent even the smallest of obstacles. A number of sensors placed in various spots around the shuttle provide both short and long-range vision, removing blind spots and enhancing safety.