Understanding Autonomous Technology: The Basics of Self-Driving Cars


The idea of self-driving cars used to be hypothetical, but with high-profile companies like Google, Apple, IBM, Uber and more making headlines for their autonomous efforts, it’s beginning to feel like an inevitability. With the reality of roads becoming the exclusive domain of artificial intelligence moving closer, it’s easy to feel confused and overwhelmed. Here are some basic facts about self-driving cars.

The difference between semi-autonomous technology and true self-driving vehicles

The distinction between today’s advanced semi-autonomous vehicles and legitimate self-driving cars may seem minute, but in reality, it is significant. Simply put, a self-driving car is a vehicle that can get from point A to point B without any human intervention. These vehicles will never need a human driver to operate safely and efficiently.

Conversely, the semi-autonomous technology found in many vehicles currently on the road is either meant to aid the driver or requires direct input, even if that input is remote. For instance, your vehicle may include driver-assist technology like automatic emergency braking or systems that keep your car centered in a lane. Additionally, Tesla vehicles have made headlines for their ability to navigate parking lots to pick up their drivers at the curb, but these cars do so via driver input from a smartphone application.

AI writer Lance Eliot has penned several articles and books on the future of self-driving cars, and through his research has devised a scale to specify levels of vehicle automation. At level zero, a driver is required at all times to operate the vehicle. Level one introduces some driver assistance features, but a human is always in control. Level two refers to vehicles that incorporate features like adaptive cruise control or lane-change assist. At level three, the car takes over some critical functions, but the driver maintains partial control. Level four vehicles are in full control unless adverse road conditions require human input. At level five, the vehicle can operate itself under any circumstances, and there is no need for human input.

How do self-driving cars work?

There are several prominent ways that today’s fully autonomous vehicles operate. The most frequently used system is based on a combination of cameras, LIDAR, radar, and machine learning. By combining the data from all of these inputs, self-driving cars constantly monitor road conditions, proximity to other vehicles on the road, and efficient driving practices

Cameras are the most easily understood component of the system because they function almost exactly how you’d expect. Lenses observe the road in all directions and help the computer driving the car stay aware of all its surroundings.

LIDAR and radar are a bit more technical but still understandable. LIDAR stands for Light Detection and Ranging and utilizes pulses of short-wavelength light to generate highly accurate, three-dimensional maps of the environment. Radar, on the other hand, emits long-wavelength radio waves to calculate the position of surrounding objects. Lidar works better at a short range and produces more detailed data. Radar returns less specific information but can sense objects from farther away. Between the two, self-driving cars have a thorough understanding of their surroundings.

Machine learning is the process through which computers store data from things they’ve observed or experienced and use it to adjust their performance in similar situations in the future. The idea behind this is very similar to the way that humans learn to drive — by remembering past experiences and using them to improve their skills.

Who is in the self-driving car business?

At this point, it’s probably easier to list the tech companies and car manufacturers that aren’t investing in self-driving car research and development. Global corporate entities like Google, Apple and Huawei have been pushing into the industry for years, and Uber is looking to eventually replace all of its human drivers with automated vehicles. Meanwhile, numerous car manufacturers like General Motors, Volkswagen and Tesla are making a major push into the market.

Are self-driving cars really safer than human-operated ones?

The most frequently cited argument in favor of self-driving cars is that they will increase overall road safety. A study by the National Highway Traffic Safety Administration found that up to 94 percent of car accidents are due to driver error. Proponents of fully-autonomous vehicles believe that giving full control over to AI could shrink this number to zero.

While it’s true that removing human input from driving would eliminate instances of drunk, angry, or otherwise compromised driving, the reality is slightly more complicated. One of the biggest hurdles the self-driving car industry faces is the mix of human drivers and artificial intelligence as the cars start driving on America’s roads. No matter how safe or efficient these vehicles may be, it’s unlikely that the entire world will turn in their steering wheel privileges at the exact same time, which means autonomous vehicles will probably have to share the road with human drivers for the foreseeable future.

However you feel about self-driving vehicles, it’s a good idea to be educated about their benefits and shortcomings. After all, they may be here en masse sooner than you think.

Related Articles

Leave a Reply