Studies from overseas have found that drivers are responsible for between 60 to 80 percent of accidents involving cyclists. Such accidents present a problem in Canada as well. Ontario lawmakers have tried for years to reduce incidents of cyclist collisions by toughening laws. Yet, the future of collision-avoidance may lie in an innovative technology: self-driving cars.
How Can A Self-Driving Cars Avoid Cyclist Collisions?
At first, the concept may seem incongruous: how can a driverless car actually prevent collisions? The answer lies in the technology found inside these cars.
Driverless cars may soon employ a technology called Deep3DBox. This is a deep learning network capable of using a two-dimensional image of an object, such as a cyclist, and subsequently, predict its three-dimensional shape and direction.
This technology would be used in combination with information gathered from cameras, radars, high-definition street-view maps and "lidar sensing systems" - a technology that colours objects with a laser to determine vehicle distance.
Essentially, the technology is designed to recognize objects and their location in order to avoid hitting it.
Is This Technology Reliable?
The invention is not quite ready to be used yet. While the Deep3DBox scored well in terms of identification rate for bicycles, the score was not yet high enough to ensure the safety of cyclists. It also needed improvement in trying to determine the predicted route of a bicycle. But once the technology has been improved and approved, the future may look brighter for cyclist accident prevention.
Gluckstein Lawyers welcomes the exciting and continuing advances in technology that promote cyclist protection and accident avoidance.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
