The July 2022 issue of IEEE Spectrum is here!

Close bar

"Swerve Assist" Algorithm Uses Power Steering and Brakes to Avoid Collision

Korean researchers keep the driver in the loop instead of having the car provide autonomous obstacle avoidance

3 min read
"Swerve Assist" Algorithm Uses Power Steering and Brakes to Avoid Collision
Photo: Marco Maccarini/Getty Images

Think of all the discrete decisions you have to make as a driver on an empty road. Layer on all the other decisions you have to make when there are other cars on the road that you need to avoid hitting. And then factor in the steering and braking reactions that need to take place in response those decisions—all within fractions of second. When it comes to steering clear of trouble on the road, humans could really use some help from cars with the smarts to help avoid collisions.

That help might soon be at hand thanks to a team of researchers from Seoul National University and the Hyundai Motor Company, both in Seoul, South Korea, who have developed an algorithm that uses vehicles’ existing power steering and braking systems to help a driver make a quicker and more controlled lane change in order to avoid a rear-end collision with another car.

In a paper that recently appeared in IEEE Transactions on Vehicular Technology, the Korean team notes that there is a substantial amount of work being done in preparation for vehicles doing autonomous steering via computer-controlled actuators. But instead of autonomous steering angle control, the Seoul researchers’ emergency driving support (EDS) algorithm enhances what the driver does.

There are already cars on the road equipped with radars and cameras that provide continuous updates on the distance between the car and the vehicle ahead, as well as the dimensions of the lane and the lane adjacent. If your car had these, whenever the car ahead of you stopped short, the EDS algorithm would likely make up for the unavoidable delay before you acted on your instinct to slam on the brakes and steer clear. The algorithm uses the information from sensors tracking the other car's rear bumper to calculate the inverse time-to-collision and the minimum lateral acceleration needed to avoid either the right or left rear corner of the leading vehicle.

Based on your car's speed and the distance between you and the vehicle you're in danger of rear-ending, the EDS algorithm directs the electric motor that provides a power assist for steering to increase the relative effect on turning angle that occurs when you turn the steering wheel. How much of an assist it provides is continuously updated based on the number crunching that happens as the sensors report just how close your car is coming to the point of impact. Once you've completed the initial avoidance maneuver, the algorithm will continue to assist you when you turn the steering wheel in the other direction to avoid overshooting the adjacent lane and possibly running off the road entirely. The researchers say they are still tinkering with a set of velocity-versus-impact distance combinations at which the steering assist should kick in.

The algorithm's other important feature comes in especially handy when, even if the steering wheel is turned so that the front wheels are angled as far as they’re designed to go, your car’s sensors confirm that it is traveling too fast and is too close to the car ahead to avoid an impending collision. In those instances, the algorithm applies the brakes to the two inside wheels, which has the effect of further increasing the amount of yaw your vehicle experiences. When you turn the steering wheel the other way to right yourself, the braking is switched to the other two wheels.

The researchers reported that, in virtual test track simulations of avoidance maneuvers around a stopped vehicle when the trailing car was 18.5 meters away and traveling at 80 kilometers per hour, cars not equipped with the algorithm could not avoid a collision. Cars with power steering support avoided crashes by an average distance of 6.5 centimeters. Those benefiting from both power steering support and differential braking missed the stopped vehicle by an average of 27.3 cm. The Seoul team also notes that, at 80 km/h, the EDS algorithm shortens, by more than a meter, the minimum distance at which an avoidance maneuver can successfully prevent a crash.

The Conversation (0)

Self-Driving Cars Work Better With Smart Roads

Intelligent infrastructure makes autonomous driving safer and less expensive

9 min read
A photograph shows a single car headed toward the viewer on the rightmost lane of a three-lane road that is bounded by grassy parkways, one side of which is planted with trees. In the foreground a black vertical pole is topped by a crossbeam bearing various instruments. 

This test unit, in a suburb of Shanghai, detects and tracks traffic merging from a side road onto a major road, using a camera, a lidar, a radar, a communication unit, and a computer.

Shaoshan Liu

Enormous efforts have been made in the past two decades to create a car that can use sensors and artificial intelligence to model its environment and plot a safe driving path. Yet even today the technology works well only in areas like campuses, which have limited roads to map and minimal traffic to master. It still can’t manage busy, unfamiliar, or unpredictable roads. For now, at least, there is only so much sensory power and intelligence that can go into a car.

To solve this problem, we must turn it around: We must put more of the smarts into the infrastructure—we must make the road smart.

Keep Reading ↓Show less