When will autonomous vehicles become a reality?

Of the six levels of vehicle autonomy, Level 1 automation like cruise control has become standard in most vehicles, while Level 2 such as auto parking has also become possible. But perfecting levels past that is still a tall order

06 July, 2024, 08:25 am

Last modified: 06 July, 2024, 08:27 am

Graphic: TBS

“>

Graphic: TBS

Picture this: It is the ’80s and your eyes are glued to BTV as it airs Knight Rider. Michael Knight, our dashing hero, is cruising down the highway in his sleek, black Trans Am. But this is no ordinary muscle car. This is KITT (Knight Industries Two Thousand), a car capable of driving by itself, chatting with its driver, and performing all manner of high-tech wizardry. Fast forward to today, and the burning question is, when will the dream of fully autonomous vehicles, akin to KITT, become a reality?

To be fair, the dream of self-driven cars probably started way earlier than the ’80s, perhaps when cars became a part of life in industrialised nations in the 1920s. The first ever invention in the field of advanced driver assistance systems (ADAS) that has been commercialised for mass production is ‘cruise control’ — a feature now standard in most cars — by Ralph Teetor.

Although it does not technically fit with the idea of an autonomous car, at least one can get the gist of it. Yet, some may disagree since it only maintains the speed and requires constant human input to control the steering and the brakes.

However, according to the Society of Automotive Engineers (SAE), there are six levels of vehicle autonomy ranging from 0 to 5. Level 0 represents no automation, where the driver controls all aspects of driving.

Keep updated, follow The Business Standard’s Google news channel

Level 1 includes driver assistance features like adaptive cruise control or lane-keeping assistance, but the driver must remain fully engaged. Level 2, known as partial automation, allows the vehicle to handle two or more tasks simultaneously, such as steering and acceleration, but still requires driver supervision. Level 3, conditional automation, permits the car to manage most driving functions under certain conditions, though the driver must be ready to take over when necessary. Level 4, high automation, enables the vehicle to operate without human intervention in specific environments and conditions, like urban areas or highways. Finally, Level 5 represents full automation, where the vehicle can drive itself anywhere, anytime, under any conditions, without any human input.

Waymo. Photo: Reuters

“>

Waymo. Photo: Reuters

Level 1 automation like cruise control and lane keeping have become standard in most vehicles and assists that require constant driver engagement. Some forms of Level 2 automation have also become possible such as auto parking with the help of cameras and sensors. However, stepping up to the next level, that makes the car take charge of the whole driving part, is yet to be normalised in mass produced cars. That didn’t stop companies from making their attempts though.

The true hands-free Level 3 driving experience was brought to the general population through the hands of Elon Musk’s Tesla Autopilot System in 2013, with the Enhanced Autopilot (EA) package being offered as an optional extra with every Tesla Vehicle sold since 2016.

In an era where artificial intelligence (AI) is being adopted everywhere, vehicles are not left behind. Through machine learning, vehicles from manufacturers like Honda, Audi and BMW are learning to adapt to the conditions of any given area, analyse this data in real-time, enabling the car to make complex driving decisions.

Termed as a semi-autonomous driving system, it combines advanced hardware and software to assist with steering, acceleration, and braking. Utilising a suite of cameras, ultrasonic sensors, and radar, EA can navigate highways, change lanes, and maintain a safe distance from other vehicles.

Other brands, which are not particularly car makers, are offering driverless ride services. These self-driving vehicles are termed robotaxis and have Level 4 automation, requiring no human intervention in the situations they are being driven at.

For example, in 2009, Google attempted to autonomously drive over ten 100-mile routes. In 2016, it merged with Waymo, an autonomous driving technology company, to make it a subsidiary of Alphabet, and renamed its self-driving project Waymo.

Waymo’s self-driving cars use a combination of lidar, radar, and high-resolution cameras to create a detailed 3D map of their surroundings. These sensors work together with advanced machine learning algorithms to detect and interpret road signs, traffic lights, and obstacles, enabling the vehicle to make real-time decisions.

Additionally, Waymo’s software processes vast amounts of data from millions of miles driven both virtually and on real roads, continuously improving its driving performance. With its fleet of fully autonomous taxis already serving passengers in Phoenix, Arizona, Waymo is on track to make the vision of driverless transportation into reality.

Cruise. Photo: Reuters

“>

Cruise. Photo: Reuters

Another company, Cruise, founded in 2013 by Kyle Vogt and Dan Kan, quickly rose to prominence with its cutting-edge autonomous vehicle technology. Initially selling autonomous driving kits that could be fitted to the Audi A4 or S4, it was soon acquired by General Motors (GM) in 2016 to enhance the automaker’s own autonomous vehicle initiatives. This merger combined Cruise’s innovative software and sensor systems with GM’s manufacturing prowess, accelerating the development of driverless cars.

Amazon has also joined the autonomous driving club by beginning tests for its own robotaxi unit Zoox in Austin and Miami. Zoox will deploy a fleet of retrofitted Toyota Highlanders in the business and entertainment districts of the two cities, but will have human drivers during the testing phase for safety.

However, as fancy as automation may sound, these vehicles are still far from being perfect. One article stated that “since May 2022, Tesla vehicles accounted for nearly 70% of reported crashes involving advanced driver-assist systems”. Besides, according to Fortune, The National Highway Traffic Safety Administration (NHTSA) in the US has received 22 reports of Waymo’s vehicles being involved in crashes or violating traffic rules last Tuesday alone.

The NHTSA has also been monitoring and pressurising several companies which were found to have been involved in routine crashes with their autonomous driving services including Honda, Ford, General Motors, and even Tesla.

Despite hurdles such as unexpected situations, traffic jams and wide variation in road conditions across the world, advancements in autonomous driving technology is far from slowing down. Now in an era where Artificial Intelligence is being adopted everywhere, vehicles are not left behind. Through machine learning, vehicles from manufacturers like Honda, Audi and BMW are learning to adapt to the conditions of any given area, analyse this data in real-time, enabling the car to make complex driving decisions.

Zoox. Photo: Reuters

“>

Zoox. Photo: Reuters

However, with Level 3 adaptation just becoming a reality, the journey to higher levels of autonomy becomes increasingly challenging but also more exciting. Level 4 autonomy, or high automation, is expected to become more prevalent by 2030.

Making this possible will also require advancements in connectivity technology like 5G adoption, 6G and beyond. Companies like Waymo and Cruise are at the forefront of Level 4 technology, already deploying autonomous taxis in limited areas.

The ultimate goal, Level 5 autonomy, represents full automation where vehicles can navigate any environment without any human input. Achieving this level will require significant advancements in artificial intelligence, machine learning, and sensor technology. Vehicles will need to be capable of handling all driving scenarios, from complex city traffic to rural roads in adverse weather conditions. While Level 5 autonomy remains a distant goal, likely several decades away, ongoing research and development are steadily pushing the boundaries of what is possible.

Source link