AD levels evolution: zooming in at functionalities and safety
Imagine traveling without steering your vehicle’s wheel. With the automated driving system engaged, you could travel for many kilometers without watching the road. What would you do? Enjoy the view, read a book, watch a movie, play some video games, or relax and nap, waking up right before you have reached your destination. In the not-so-distant future, all this will be possible. Yes, those futuristic scenes of George Jetson ‘driving’ his vehicle with his legs lying down on the car’s dashboard, and without moving a finger, are right behind the corner. But dreamer, hold on a second because before this happens, the automotive industry must solve some “minor” issues.
What driving aid do cars provide nowadays, and what scenario seems ahead of us? The automotive industry has cataloged automated driving into five advanced driver-assistance systems (ADAS): ADL2, ADL2+, ADL3, ADL4, and ADL5. For each case, the industry is currently in the process of improving features and overcoming challenges.
The operational design domain of functionalities increases when moving from ADL2 to ADL2+. This progression requires the availability of features, eventually everywhere and all the time. ADL3 introduces a paradigm shift from human to machine responsibility. A machine driving autonomously in a specific street section (for instance, a highway) requires a meticulous and trustworthy geofencing operation. ADL4 and ADL5 promise a mobility service revolution with robot taxis, automated shuttles, and buses capable of driving people across multiple environments.
For all these cases, precise localization is fundamental. One main component is the GNSS-based positioning. u-blox, a leader in this technology, provides quality, reliable, and high-performance solutions for all cases and architectures. Its solutions are designed to perform under challenging scenarios, with upgradability for new use cases and pre-certification when necessary.
Use cases:
ADL2/L2+:
ADL2 and ADL2+ help drivers to handle different driving scenarios autonomously. These embedded systems let people travel with their hands off the wheel; ‘eyes off’ is still a wish, yet this system assists drivers in their journey. Although vehicles make decisions independently, the responsibility lies with the driver, whose awareness should always be on, even if the hands do not direct the wheel all the time. Features such as lane keep assist, automatic cruise control, or traffic jam pilot can be enabled in specific geofenced zones. ADL2+ extends the operational design domain of ADL2 features for eventual support everywhere.
ADL3:
ADL3 promises car passengers to keep their hands and eyes off the road ‒ the automated driving system breakthrough. The driver is no longer responsible for decisions. When the feature is on, the driver can be ‘eyes off’ and ‘hands off.’ Based on reliable detection, the system should activate certain functionalities in precise geofenced areas with specific zone/environmental conditions. The real challenge for deploying this system lies in technological improvements and regulations that must be in place; enabling a machine to make decisions implies a liability shift that takes time to implement. Considering this, while car OEMs have started to deploy the first ADL3 fleet, large-scale deployments of ADL3 for passengers should increase considerably towards the decade’s end.
ADL4 & ADL5:
Vehicles with ADL4 and ADL5 systems will move autonomously without a driver or human supervision in restricted areas or other scenarios. This level of autonomy corresponds to mobility as a service and automated mobile robots, promising a revolution in mobility services, where robot taxis, automated shuttles, and buses, among other devices, will hit the streets. Thanks to this technology, mobility services are expected to reduce costs per km, leading to democratization. These systems can also be used for robots delivering packages to users, driving either on streets or sidewalks. Long-haul trucks present another potential business case for ADL4 as these systems could enable them to autonomously travel long distances on highways.
The GNSS sensor is essential for automated driving
Depending on the AD level, drivers or vehicles need to establish their location in relation to the surrounding space. This is fundamental for deciding the best maneuvers according to circumstances. Typically, location functions rely on several inputs (camera, radars, HD-MAPs, IMU, GNSS, and other technologies). However, the GNSS is the only sensor enabling absolute positioning with decimeter accuracy. It is also available in conditions when other sensors might be compromised.
GNSS enables not only accurate geofencing but also increases the operational design of assistance or automated functions through improved availability and accuracy of the location function.
GNSS-based positioning functionality can also satisfy safety goals for deploying highly automated driving solutions safely.
u-blox offering in ADAS
The myriad of vehicles’ architectures, features, sensors, etc., requires extensive positioning solutions. u-blox’s portfolio is broad, enabling the adaption to multiple OEM E/E architectural designs. They can also be integrated into different DCUs (ADAS, Infotainment, and Telematics) or ECUs designed for positioning/location functionality. The positioning solutions include hardware components with GNSS receivers, software components including dead reckoning (DR), RTK algorithms, cloud-based assistance, and correction services for dm-level accuracy, when necessary.
The requirements differ depending on AD levels. Thus, adapting to requirements levels and usage of the vehicles’ positioning information means offering a spectrum of solutions. These encompass multi-band solutions to achieve sub-meter level accuracy, multi-band solutions with SSR/OSR correction service for high accuracy, and pre-certified safe solutions in accordance with ISO 26262.
Over the last 25 years, u-blox has proven its strength in this challenging market, supplying automotive OEMs, Tier-1s, and device manufacturers ever since. The company has a proven track record in positioning solutions with a presence in more than 60% of the (sizeable) market.
Standard precision solution. It encompasses a single-band or multi-band GNSS receiver as a module or chip-down solution, with fully integrated sensor fusion and dead reckoning algorithm. This guarantees fast time to market, full pre-validation, and solution integration.
High precision solution. A solution that includes multi-constellation, multi-band GNSS receiver with integrated high precision algorithms, L-band receiver, and GNSS correction service to achieve dm-level accuracy (with continental coverage, optimized user plane bandwidth usage, and open SPARTN data format).
