The transition from human driven vehicles to autonomously driven vehicles requires an entirely new approach to system design. Gone are the fail-safe actuation systems of yesterday; in are fail-operational actuation systems that enable driver-less vehicles to safety operate unrestricted on the world’s roads.
Since electrification of vehicles started in the early 2000’s, the automotive sector has become one of the most rapidly changing markets in the world.
The following decade was marked with failed pledges, but now the promises of autonomous mobility are starting to be realised. The wide deployment of autonomous vehicles across the world may well be the greatest and final step towards a world where automotive vehicles become the norm.
This presents a massive opportunity for the development of new technology. However, for legacy tier 1 suppliers, this poses an existential threat. Rapid change is at odds with their business models; requiring new ways of thinking, quick development cycles and a deviation from the norm. The oligopoly of legacy tier 1 suppliers cannot embrace change on this scale on terms other than their own.
Legacy tier 1s continue to develop steer-by-wire products for human driven vehicles, claiming their concepts are new. However, employing a rack and pinion in a steering system that requires no mechanical connection to a steering wheel is absolutely not new.
In contrast, Chassis Autonomy brings an entirely new approach to automotive product design, and gone is the legacy approach of ‘carryover modified’. Our CS1 system eliminates the rack and pinion concept, deletes the parallel axis powerpack design and applies a ground-up approach to designing a critical actuation system that is wholly optimised for safety and function.
“Conventional fail-safe actuation systems cannot assure safe vehicle operation when used in autonomous vehicles because the very basis of the redundancy concept that they are designed upon (the human driver) is no longer present.
Fail-operational systems on the other hand are designed to ensure functionality even in the event of a system fault or failure. By definition, these systems must exceed current industry best practice for functional safety (ASIL-D ISO 26262) to assure safe operation, providing the confidence and credibility demanded from regulators and the public for the wider acceptance and deployment of autonomous vehicles on the world’s roads”
– Thomas Li, Co-founder and CTO Chassis Autonomy
In a world where autonomous vehicles operate without a human driver, safety is paramount. Therefore, a new approach is required to ensure the safe operation of autonomous vehicles.
Conventional thinking and ways of ensuring safety in human driven vehicles are no longer applicable. Fail-operational systems are a fundamental requirement to make sure that critical systems are able to continue to operate and function even in the event of a fault.
For example, Chassis Autonomy’s CS1 steer-by-wire system employs a unique mechanical, electrical and software architecture to ensure fail-operational capability and continued functionality even in the event of a single point failure (sensor, power, communication or other fault).
Put simply, CS1 ensures that the autonomous vehicle maintains steering functionality even in the event of a system or vehicle fault, enabling the continuance of a journey that would otherwise require the vehicle to perform a potentially dangerous emergency stop.
For autonomously driven vehicles to operate unrestricted and gain mass global approval, fail-operational systems are not only a fundamental requirement to ensure safety, they are now becoming a legal requirement.
Regulations in Europe and other parts of the world will soon outlaw the current practice of stopping in-lane, in the event of a system or vehicle fault. Simply put, the use of fail-safe systems will not be possible and only the use of fail-operational systems that guarantees operation will be permitted.
Functional safety, specifically with regards to ISO 26262 covers the electrical, electronic and software design elements of automotive systems. It is derived from industry best practice and aims to ensures the safe operation of such systems in the event of a fault or failure, where the complexity of design and level of redundancy is proportional to the risk and hazard posed by a failure of those systems to function normally.
ISO 26262 was released in 2011 when autonomous vehicles were little more than a distant dream, with the human driver very much fore and centre. Fast forward to today and it’s clear that safety must take an even greater focus when the human driver is removed from the loop; and whilst ISO 26262 is still relevant, a step change is required to assuring the safe operation of critical systems for autonomous vehicle use.
Chassis Autonomy’s fail-operational systems are designed to be inherently safer than conventional fail-safe systems. Our systems are designed from the ground-up, upon a principal collaborative approach between our functional safety experts and principal brake and steering system engineers, to deliver products that exceed the current state-of-the-art in functional safety, ASIL-D.
Furthermore, as vehicles become more and more connected, the security of these critical systems becomes paramount to preventing security attacks that could take over control of the vehicle. Our systems are designed to comply with the cyber security requirements of ISO 21434, UNECE R155 and UNECE R156 (with regards OTA updates).
Chassis Autonomy’s class-leading technology is unique and a valuable asset to the company. We therefore have 13 patents pending to protect our intellectual property and de-risk the development of our products. These patents add value to the company, ensuring our technology cannot be replicated by other companies.