A Look at the Autonomous Vehicle Object Avoidance Subsystem

One of the most discussed and debated questions in philosophy is, “if a tree falls in the forest and no one hears it, does it make a sound?” George Berkeley, an Anglican Bishop, writer and philosopher, is credited as the first to put forth this query. It seems that the most widely accepted answer is no. If there is no perception, then there is no occurrence. In other words, perception is reality, while the uninformed listener is oblivious to whether the event occurred or not.

As a passenger in an autonomous vehicle or AV, you rely on the object avoidance system to not only prevent collisions but also keep you oblivious to potentially harmful occurrences. The ability to identify hazards and perform necessary avoidance actions is the stalwart upon which driverless car technology is based. Let’s lift the hood on this important autonomous vehicle system and then examine industry trends in object avoidance systems.

The Autonomous Vehicle Object Avoidance Subsystem

Today, all vehicles can be classified based on the sophistication, capabilities, and level of driving autonomy provided by the electronics and PCBs. For most of the levels, which range from 0 for total driver control to 5 for total vehicle control, some degree of automatic control is assumed by the vehicle. This may include common functions, such as cruise control and automatic braking, or may include multiple functions that provide more sophisticated control. Object avoidance requires multiple functions or capabilities to be effective.

AV Object Avoidance Subsystem Elements

Object avoidance subsystems are one of the major advanced driver assistance systems (ADAS) and may include some of the following elements or systems:

• Light detection and ranging (LIDAR)

LIDAR is a technology used to estimate long-range distances.

• Ultrasound

Sound waves are typically used to gauge distance, similar to radar systems.

• Short-range radar sensing

Short-range radar is used for distance estimation near to the vehicle.

• Sensors and cameras

Sensors and cameras may be located at the front, sides, rear, and interior of the vehicle. These are mostly used to identify potential collision obstacles. However, cameras and sensors are also integral for advanced AV technology, such as blind spot detection, street sign recognition, and pedestrian crossing monitoring.

• Collision warning system (CWS)

The CWS is a mid-level AV ADAS as it depends upon a driver to respond to an audible warning.

• Automatic emergency braking system (AEBS)

Automatic braking coupled with an obstacle detection method is an advanced safety feature that high-level AVs must possess.

AV Object Avoidance Subsystem Operation

The object avoidance systems for autonomous vehicles usually operate in one of the three ways listed below, depending upon the level of autonomousness.

Operation(s)	Autonomous Vehicle Level(s)
Warning only	0
Action only	1, 2, 3, 4, 5
Warning + Action	2, 3, 4, 5

The lists above present the possible levels of autonomy for the three object avoidance system operation types. Currently, there are vehicles on the road at all levels between 0 and 4; however, the trends in electronics and PCB development point toward a future with fully autonomous Level 5 vehicles on our highways.

Trends in Autonomous Vehicle Object Avoidance

In order for fully autonomous vehicles to achieve the safety, reliability, and performance required to not only navigate the roadways without incident but also lower the apprehension of the public, autonomous vehicles must continue to refine the following types of systems:

• Communications

Vehicle control through communications, such as the global positioning system (GPS) has been around for nearly two decades. For example, OnStar, which provides security and other services, remotely was launched in 2001. The ability for vehicles to communicate quickly and effectively internally, between vehicles and as part of the larger Internet of Things (IoT) systems, will be critical to advance object avoidance to the reliability necessary for Level 5 autonomous vehicles.

• LIDAR

Although some radar systems have a greater range, the precision of LIDAR is expected to be a central part of Level 4 expansion and Level 5 implementation.

• Control

Collecting, interpreting, and predicting the best action to take is the function of the AV control system. It is predicted that machine learning and training algorithms will play a significant role in the further development of these systems.

• CWS

Collision warning systems will continue to be an object avoidance subsystem staple; however, these will need to be combined with actions to avoid the collision. An example is a dynamic braking system (DBS) where the CWS is combined with the capability to adjust the braking as necessary depending on the situation.

A future with fully autonomous, Level 5, vehicles on the road appears highly likely in the near future. The promise of virtually removing the risk of collisions, which too often end in fatalities, from the roadways is tantalizing. At the heart of achieving this objective is the quality and reliability of the PCBs and electronics that comprise autonomous vehicle object avoidance subsystems.