Could AI Help Reduce Bike Accidents?
When we think of artificial intelligence, we often think of technologies like self-driving cars, smart home systems and various smartphone apps. However, a software company wants to use AI to help prevent bicycle accidents.
According to the U.S. Department of Transportation, 818 bicyclists were killed on U.S. roads in 2015, which is the highest number since 1995. Meanwhile, 45,000 bicyclists were injured. Half of all fatal crashes occurred during daylight hours.
While protective gear like bicycle helmets and reflective clothing can help reduce accidents, it clearly can’t eliminate all fatal collisions between motor vehicles and bicycles. To address the issue, Tome Software is starting a trial that uses AI to prevent several types of bicycle accidents. The Detroit-based company is teaming with Trek bikes and the University of Michigan’s TechLab to develop AI that will be capable of alerting drivers and cyclists of an impending collision. The software will particularly focus on preventing cars and trucks from hitting bikes from the side or behind, which are the most common types of bike/car accidents.
How will this technology differ from collision prevention systems already available on many cars and trucks? According to Venture Beat, Tome intends to use several data sets, such as road characteristics, time of day and previous crash data from the area, to teach AI to produce more accurate warnings to riders and drivers. The hope is that this will prevent false-positive alerts, which can annoy people and encourage them to tune out warnings. The goal of the research is to create software that will only alert riders and drivers when a cyclist is truly in danger of being struck.
According to TechCruch, Tome will begin their project by using AI to map the most dangerous areas of a road. The company will then figure out the best way to share the information with cyclists and drivers.
Meanwhile, Trek is confident that its partnership with Tome will improve safety for cyclists.
Signal founded Tome Software in 2014. The company develops connectivity solutions for large corporations. Its first product was the HumanScale desk, which helps office workers keep fit by telling them when to sit and stand at their desk. The company’s bicycle AI project is just now entering the research phase.