Autonomous driving is the automotive industry’s buzzword these days. With non-automotive companies like Google and even China’s Baidu threatening to muscle into the automotive space by leveraging on their knowledge in artificial intelligence to build self-driving cars, car companies like Toyota are busy working on their own alternatives.
While Tesla proudly markets (semi) autonomous driving features in a rather misleading ‘Auto Pilot’ name (already blamed for at least one death), Toyota, being a traditional car maker have repeatedly express caution over pushing a technology that is not yet fool-proof to the public, and the dangers of giving a wrong impression to the public.
At the 2015 Tokyo Motor Show, President Akio Toyoda made a subtle reference to Tesla’s practice of rolling out relatively unproven Auto Pilot the public, and warned that all it takes was for only one highly publicized accident and progress for self-driving cars will come to a halt.
“The technological advancement would stop suddenly. We have to have a very long-term perspective,” he said.
Now, Toyota has published a white paper – Automated Driving At Toyota, outlining the company’s approach towards automated driving.
The term ‘automated driving’ was intentionally chosen over the more popular ‘autonomous driving’ term. In Toyota’s opinion, the distinction is important because many autonomous vehicles aren’t really autonomous, meaning that the car is not realistically capable of absolving the human driver from any responsibility or supervision.
Toyota would only refer to cars that are 100 percent capable of a full range of dynamic driving tasks as autonomous (meets SAE J3016 Level 5 requirements). Short of that, it will simply refer to it as automated vehicles.
“Care using these terms is important, as their application to vehicles in the market may impact consumer expectations and understanding about how those vehicles perform. As we implement these technologies in passenger vehicles, we believe it is more appropriate to describe accurately, or to use terminology that suggests, the actual function the vehicle can perform,” said Toyota.
The Society of Automotive Engineers (SAE) has defined automated (or autonomous) driving vehicles in several different levels of capabilities (below). While Toyota acknowledges the SAE J3016 standard for ease of industry-level discussions, it doesn’t agree that the SAE standard adequately covers the complexity of the task ahead.
For example, Toyota finds the definitions for Level 3 particularly difficult to agree to. A Level 3 system is not yet capable of a full-range of driving tasks. However the driver is not required to supervise the driving but yet is expected to respond fast enough when the driver is prompted to take over driving responsibilities, should the vehicle encounters a situation it can’t manage.
It also disagrees to SAE’s taxonomy of using the system’s ability to handle different driving conditions to differentiate between Level 4 and Level 5, as this complex metric needs to be applied to all levels of automated driving, and in the real world, manufacturers develop systems that combine several different levels, depending on driving conditions, simultaneously.
Toyota’s Alternative – A Guardian And A Chauffeur
Instead, Toyota’s internal research on automated/autonomous driving is split along a different, simpler hierarchy – Chauffeur mode and Guardian mode, both relying on similar hardware. The software and hardware used for Guardian mode function serves as the basis for the more advanced Chauffeur mode.
Basic level Guardian mode functions are already available in existing Toyota/Lexus vehicles – features like Lane Keeping Assist, Autonomous Emergency Braking. More advanced Guardian mode functions of the future will see that no human-driven vehicle will cause a crash, regardless of any error made – yes that’s Toyota’s bold claim.
It even added that the vehicle will be able to steer away from most accidents caused by other vehicles, or other external factors, the keyword here being ‘most.’
In fact, Toyota tempered its bold claims against crash-proof Toyota/Lexus vehicles of the future by adding, “Importantly, driving environments can be extremely complex and difficult, and no automated driving system – regardless of how capable it may be – is likely to prevent crashes entirely. A fundamental question yet to be addressed is “how safe is safe enough?” The answer will depend on government regulation, liability risks, societal acceptance, and what is technically possible.”
This is quite different from the Volvo Car Group’s Vision Zero philosophy, which aims to achieve zero fatality or injury in any new Volvo car by 2020, regardless of driving conditions.
Building on the Guardian mode is the more advanced Chauffeur mode. At its highest development level, Chauffeur mode offers true, fully autonomous driverless cars.
The Toyota Research Institute’s Platform 2.1 is the company’s latest proof of concept for automated/autonomous driving features. Built on an outgoing generation Lexus LS, this is a further development of Platform 2.0 that was showcased in March 2017. The upgraded model features a new hi-fidelity LIDAR (Light Detection and Ranging) that is the first to be dynamically configurable, meaning that the LIDAR is able to concentrate on specific areas where sensing is needed most, just like a human eye.
Inside the cabin, there is a secondary set of drive-by-wire steering wheel and pedals, to study acceptable nuances when handling controls between human to computer and vice-versa when faced with challenging situations that the computer cannot handle – which as explained earlier, is one of the most difficult challenges in creating an autonomous driving cars that is acceptable by the masses.
Platform 2.1’s cabin also includes a screen that shows a point-cloud representation of everything the car’s computer is seeing, part of a study to solve the problem of how to increase a driver’s situational awareness when control is being passed to the computer.
Toyota Is Not Replacing The Driver, And That’s Good News
The biggest difference in Toyota’s approach towards autonomous cars compared to other manufacturers is that Toyota has no intention of replacing the driver.
This contrasts greatly from recent autonomous driving concepts demonstrated by Daimler and Audi, several of which don’t even have a steering wheel, and if their concepts are a reflection of the German industry’s plans for the future, then every car of the future is going to look very similar, all of them lounge/conference room on wheels with not even a steering wheel, all with the same electric driving characteristic.
Instead of replacing the driver, Toyota seeks to build a more ‘friendship’ style relationship between the car and a driver – each looking out for the other, a relationship that Toyota refers to as Mobility Teammate Concept (MTC).
Toyota recognizes that there are certain things are just better done, worth enjoying by humans, while other more mundane repetitive but complex tasks are better done by computers.
Toyota firmly believes that the humans must be given to choice to either drive themselves or to allow a computer to do it, not surprising considering that Akio Toyoda himself is a keen enthusiast driver and has been whipping his engineers to make Toyotas ‘fun to drive again.’ Emotional designs like the all-new 2018 Toyota Camry, C-HR and Lexus LC are all a preview of things to come for Toyota.
“MTC is a philosophy built on the belief people should have choices. Rather than removing humans from any engagement with their own mobility, this allows people to enjoy the fun and freedom of driving when and if they choose, while also benefiting from the capabilities of automated driving when they wish,” said Toyota.
By using connected systems and cloud-based technology, Toyota is also taking advantage of the wide range of skills in the pool of human drivers, allowing the experience of every individual driver to improve the system. In other words, the computer will continuously learn from better skilled, smoother human drivers, many of whom are likely to want to continue driving themselves.
Elderly And Disabled Drivers To Find New Purpose As Pioneers
Like any new technology, automated/autonomous driving cars will be too expensive in the initial stages, and interest from the public are likely to be low. Without widespread usage, data gathering will be difficult, slowing the system’s ‘learning’ process.
Toyota’s solution is to roll out automated/autonomous driving not through private buyers, but via partnership with various companies to offer mobility solutions to people with limited ability to drive – the disabled and the elderly – a service Toyota describes as Mobility As A Service (MaaS).
Challenges To Autonomous Driving
The white paper also outlines issues that needs to be addressed before rolling out autonomous driving cars, most of which related more to legal, cultural, social norms, city planning rather than technical.
“One key and unique success factor for automated driving is how we collaborate and cooperate with various stakeholders who are not always actors in the traditional auto industry. To realize and popularize automated driving, we need to appreciate their expectations. Industry collaboration in non-competitive areas that act as a foundation of automated driving systems and vehicles, such as infrastructure or social systems, is an effective way forward.”
More details for the geeks among you can be found at automatedtoyota.com.