Over the past decade, the United States has seen an explosion of autonomous vehicle technology that has swept across the auto industry. This wave of progress encompasses all aspects of computer technology, software engineering, and thought-leaders from major automakers like Tesla, BMW, Ford, Audi, and even Google. While many have only started hearing about autonomous technology recently, self-driving car research has been going on now for over 45 years.
One of the earliest research publications on autonomous vehicle technology can be found in an article IEEE Spectrum from 1969. In the featured article, lead engineers Robert E. Fenton and Karl W. Olson hypothesized that the future of automated vehicles would rely on “smart infrastructure” that would guide the cars on roadways.
See also: Will vehicle-to-vehicle communication in new cars save lives?
Since that article was published, we have witnessed unprecedented advancement in computer technology and information systems. These advancements now allow for faster computers that are smaller and lighter than imagined in the late 1960’s. As a result, autonomous vehicles now rely on onboard technologies and state of the art computers to observe and process their environment.
While the technology used has made leaps and bounds, the ability for computers to understand their surrounding and make decisions based on relevant information has also improved. Artificial Intelligence (AI) plays an integral role in the progression of self-driving vehicles on public roads.
AI: the brain of autonomous vehicles
Just like a human, self-driving cars need to have sensors to understand the world around them and a brain that collects, processes and chooses specific actions based on information gathered.
The same goes for self-driving cars, and each autonomous vehicle is outfitted with advanced tools to gather information, including long-range radar, LIDAR, cameras, short/medium-range radar, and ultrasound
Each of these technologies is used in different capacities, and each collects different information. However, this information is useless unless it is processed and some form of action is taken based on the gathered information.
This is where Artificial Intelligence comes into play and can be compared to the human brain, and the actual goal of Artificial Intelligence is for a self-driving car to conduct in-depth learning.
In a recent interview, Sameep Tandon, CEO and co-founder of Drive.ai, explains that “deep learning is the best enabling technology for self-driving cars.” He goes on to explain that “you hear a lot about all these things on a car: the sensors, the cameras, the radar, and LIDAR. What you need are the brains to make an autonomous car work safely and understand its environment.”
Artificial Intelligence has many applications for these vehicles; among the more immediate and obvious functions:
- Directing the car to a gas station or recharge station when it is running low on fuel.
- Adjust the trip’s directions based on known traffic conditions to find the quickest route.
- Incorporate speech recognition for advanced communication with passengers.
- Eye tracking for improved driver monitoring.
- Natural language interfaces and virtual assistance technologies.
Helping autonomous cars learn from each other
At its core, Artificial Intelligence is a complex algorithm that mimics how the human brain learns. Instead of hard-coding an autonomous car with thousands of “If-Then” statements, software engineers create an algorithm that outlines to the car’s onboard computers various examples of what is right, wrong, safe, and unsafe for the car to perform.
This type of approach to automotive engineering may seem counter-intuitive, but in reality, artificial intelligence algorithms are the only solution to the dynamic driving conditions of public roads.
There is no way for engineers to hard-code every possible variable or situation a car may face in a daily drive. Instead, engineers rely on the ability for the autonomous car to collect information and then process it through the fluid Artificial Intelligence algorithm.
The real power of this approach is realized because autonomous cars have one advantage that human drivers don’t have; self-driving cars have the ability to share their experiences and readings with other cars instantaneously.
Information and situations encountered by autonomous cars along every mile driven are shared with other vehicles so that each computer can adapt its algorithm to the environments faced by other vehicles.
This type of shared experience and active learning creates a situation where autonomous cars, through Artificial Intelligence algorithms, can improve their ability to react to situations on the road without actually having to experience those situations first-hand.
The software for smarter cars tomorrow?
Self-driving cars are rapidly evolving as we see unimaginable innovation in hardware, software, and computing capabilities. However, as we progress toward advanced automobiles, one of the limiting factors restricting the growth of this field is Artificial Intelligence and machine learning.
Unless autonomous cars can interpret the many types of objects and situations surrounding them, they can’t make adequate decisions. Instead of developing millions of rules, a sophisticated learning algorithm needed to develop and standardized across the industry.
The entire self-driving car industry will suffer if only specific makes and models of autonomous cars are fitted with proper Artificial Intelligence software. Because, while not necessarily accurate, our society views all autonomous cars as a single entity.
If a Tesla causes an accident or an Uber speeds through a red light, our society attributes that error to all autonomous car technology.
This means that not only does the future of autonomous cars depend on advanced Artificial Intelligence algorithms, self-driving cars also rely on the standardization of that algorithm across all autonomous vehicles. Without this shared technology, we can’t expect our society or policy makers to accept autonomous cars on public roads on a wide-scale.