Global
The automotive industry is racing to deliver highly anticipated autonomous driving vehicles without human intervention.
On the journey to a fully autonomous vehicle, companies have explored many paths to SAE L5 (Society of Automotive Engineers Level 5).
However, they face some autonomous driving challenges with Level 3 (L3) due to the complexities of combining automation technology with human involvement.
In this article, Jean-Paul de Vooght, AVP of Client Solutions, explores the autonomous vehicles challenges and how to overcome them.
Autonomous Driving
Developing an L3 vehicle requires sophisticated hardware, software, algorithms, and an immense amount of data.
This must work together seamlessly to enable the Electronic Control Units (ECUs) to make decisions and execute them when a vehicle slams its brakes or a child wanders into the street.
Advanced Driver Assistance System
The automotive industry is actively working on fail-operational safety architectures for systems with the powertrain in the context of the Advanced Driver Assistance System (ADAS) and Autonomous Driving (AD) processing chain – from sensors to perception and decision algorithms.
This work has led to architectures incorporating hardware and process redundancy, real-time fault detection, masking, and advanced reconfiguration to sustain normal operations after a fault.
The underlying assumptions for fail-operational architectures include partial fundamental redundancy through graceful degradation, detached sensors/actuators, communications via separate paths, synchronized compute platforms, and dissimilar powertrain platforms.
Traditional Domain-Centric and Legacy-Heavy E/E Architecture
The automotive industry has long recognized that traditional Electrical/Electronic (E/E) architecture is too domain-centric and legacy-heavy.
It has likely delayed original timelines to achieve SAE L4 autonomy and is compounded by the electrification of the powertrain, which introduces additional safety concerns.
The significantly higher compute capacity implied by Automotive Safety Integrity Level D (ASIL-D) systems for autonomous driving has led to the development of more efficient multicore ECUs, hopefully reducing the count by domain and, thus, the overall complexity.
Early designs suggested a simplified vehicle model based on Adaptive AUTOSAR.
It leverages commercial off-the-shelf in-vehicle infotainment, classic AUTOSAR, and 5G/IP communication with high coordination between ECUs, one-out-of-two fail-operational systems, and metrics such as a maximum fault-handling interval used to identify a safe-state transition.
More recent efforts explore two-out-of-three fail-operational systems of multicore ECUs.
This design is widely used in avionics and consists of three fully independent redundant elements from the sensor to the actuator.
The Irony of Automation
L3 autonomous driving poses specific safety risks that are inevitable when a system relies on automation and human supervision.
The “irony of automation” refers to the fact that an automated system that requires supervision can retain the attention of a human operator.
In contrast, a system with greater autonomy will eventually lose the operator’s attention, making a re-entry into the supervision loop much harder.
The recent Boeing 737 MAX MCAS situation has added to the ongoing discussions on the balance between fully automated functions of a fail-operational system and relying on the driver or pilot to deal with emergencies—something pilots are trained for specifically.
Still, they are not part of our current driver’s license curricula.
While flying a plane may seem more complex than driving a car, achieving a self-flying airplane is easier than a self-driving car.
It is because many events could interfere with a vehicle, including pedestrians, traffic, animal crossings, and countless other interactions that drivers must be aware of.
An L3 system cannot be expected to make those decisions and self-correct accordingly without human supervision.
Machine Learning for Autonomous Driving
Developing ML algorithms suitable to meet the collision avoidance safety goals on the ADAS/AD processing chain will require continuous improvement and development. Preferably a vehicle is equipped with the requisite connectivity such that the ML algorithms can be upgraded aftermarket.
Rich, diverse, and curated training data and efficient algorithm-tuning workflows will be necessary to ensure the system performs correctly and provide an opportunity to continuously improve the algorithm to help the vehicle make better decisions.
With clever data engineering and data science to manage and optimize their ML pipeline for training autonomous driving algorithms, automotive companies can leverage L3 driving with consideration for higher levels of autonomy.
However, it will require constant innovation, perfected combinations of sensors, and transparency with consumers about the vehicle’s capability to process data and the need for the driver to remain vigilant and prepared to jump back into the supervision loop.
About the Author
Jean-Paul De Vooght (JP)
AVP—Client Solutions
JP has been a member of the Ness Solutioning team since 2012 and has previously been a systems integration consultant and solution architect in various industries. Past domain experience includes B2C e-commerce, mobile value-added services, social media syndication, and live A/V streaming. At Ness, Jean-Paul has conducted various Discover and Envision workshops in the Manufacturing and Transportation industry. He is also a member of the Internet of Things (IoT) practice core team.
