A self-driving car (autonomous car) is a vehicle that travels between locations without the assistance of a human operator. It does it via the use of sensors, radar, and artificial intelligence (AI). To qualify as completely autonomous, a vehicle must be capable of navigating to a specified location without human intervention on roads that have not been modified for its usage.
Audi, Ford, Tesla, and Volkswagen are among the companies developing and/or testing autonomous vehicles. Google’s test included traversing over 140,000 miles of California streets and highways using a fleet of self-driving vehicles, including a Toyota Prii and an Audi TT.
The operation of self-driving vehicles
Self-driving vehicle systems are enabled by artificial intelligence technology. Self-driving vehicle developers use massive quantities of data from image recognition systems, as well as machine learning and neural networks, to create systems capable of driving independently.
Neural networks discover patterns in data, which are then fed into machine learning algorithms. This data comprises pictures captured by cameras mounted on self-driving vehicles, which the neural network uses to learn to recognize traffic signals, trees, curbs, pedestrians, and street signs in every given driving environment.
For instance, Google’s self-driving car project, Waymo, integrates data from sensors, Lidar (light detection and ranging – a technology akin to radar), and cameras to recognize everything in the vicinity of the vehicle and anticipate what those things will do next. This occurs at fractions of a second intervals. These systems need maturity. The more miles driven, the more data the system can integrate into its deep learning algorithms, allowing for more sophisticated driving decisions.
The following describes the operation of Google Waymo vehicles:
- A driver (or a passenger) establishes the destination. The route is calculated by the car’s software.
- A spinning, roof-mounted Lidar sensor continuously scans a 60-meter radius surrounding the vehicle, creating a dynamic three-dimensional (3D) representation of the vehicle’s current environment.
- A sensor on the left rear tire monitors sideways movement in order to determine the car’s location in relation to the three-dimensional map.
- Radar sensors integrated into the front and back bumpers measure the distance between vehicles and objects.
- The car’s AI software is linked to all of the sensors and gathers data from Google Street View and the car’s video cameras.
- The AI mimics human perception and decision-making processes through deep learning and regulates actions in driver control systems such as steering and braking.
- Google Maps is consulted by the car’s software to get prior warning of landmarks, traffic signs, and lights.
- There is an override feature that allows a person to take control of the vehicle.
- Automobiles are equipped with self-driving capabilities.
Features of Self-Driving Car
Google’s Waymo project is an example of an almost fully autonomous self-driving vehicle. A human driver is still required, but only to overrule the system when necessary. It is not really self-driving, but it is capable of driving itself under perfect circumstances. It has a great degree of autonomy. Many of the vehicles on the market now offer a lesser degree of autonomy but still have some self-driving capabilities. As of 2019, the following self-driving features are present in the majority of production cars:
- Without the driver’s hands on the wheel, hands-free steering centers the vehicle. The driver still needs to maintain a high level of concentration.
- Coming to a stop, adaptive cruise control (ACC) automatically maintains a preset distance between the driver’s vehicle and the vehicle in front.
- When the driver crosses the lane lines, lane-centering steering intervenes by automatically pushing the car toward the other lane marks.
- Autonomy levels in self-driving vehicles
The Levels of The Car
The National Highway Traffic Safety Administration (NHTSA) of the United States defines six degrees of automation, starting with Level 0, where people drive, and progressing to completely autonomous vehicles. The following are the five degrees of automation that follow Level 0 automation:
- Level 1: An advanced driving assistance system (ADAS) assists the human driver with steering, braking, and acceleration, but not all three at the same time. A driving assistance system (ADAS) contains rearview cameras and features such as a vibrating seat warning system that alerts drivers when they stray out of their travel lane.
- Level 2: An ADAS capable of steering and braking or acceleration concurrently while the driver is fully conscious and acts as the driver.
- Level 3: In some situations, such as parking the vehicle, an autonomous driving system (ADS) can handle all driving duties. In these instances, the human driver must be prepared to regain control of the car and must continue to be the primary driver.
- Level 4: In some situations, an ADS can handle all driving duties and monitor the driving environment. In such instances, the ADS is sufficiently trustworthy that the human driver is not required to pay attention.
- Level 5: The ADS of the car serves as a virtual chauffeur, driving the vehicle in all situations. Human occupants are always supposed to be passengers and are never expected to operate the vehicle.