Protecting the people
In the wake of the Uber self-driving car that hit and killed a woman in Arizona in 2018, a lot of coverage was given to the effectiveness of the safety systems in the car and how reliable they really are. In that incident, the conclusion was that the technology wasn’t designed to detect jaywalkers, hence the impact. But, armed with this information, companies have worked to change and improve on product offerings with the intention of never having such an incident in the future.
One solution comes from Finnish city Tampere, where Tieto—a Nordic IT services and software company—has been working on extra protection for pedestrians. Working with the city, Tieto has used artificial intelligence (AI) and internet of things (IOT) technologies to automatically detect when a pedestrian is planning to cross the street at an intersection. In the event of this scenario, an alert can be relayed to automatic traffic signs and—in the future—directly to other vehicles. The work being carried out is being seen in Finland as a building block to fully autonomous vehicles.
In Tampere, an intersection traffic camera feed was connected to a cloud-based AI system that monitors vehicles and pedestrians. When the system’s algorithms detected motion from a pedestrian—i.e., the person starting to cross the road—the alert sounded.
“By enhancing existing traffic-monitoring technology with artificial intelligence, we can better identify traffic accident risks,” said Jari Torkkola, Program Director, Tieto Product Development Services. “The system monitors the movements of vehicles and pedestrians, and recognizes when a pedestrian intends to cross the street. Especially in areas of limited visibility, the system can help prevent accidents.”
The initial results of the trial have proved positive, according to the project team. Under ideal conditions (daylight, dry weather, clear visability), the system achieved 99% accuracy. Even at night the accuracy was rated at 75%.
“We had identified the most common types of accidents between vehicles and pedestrians,” explained Pekka Stenman, a traffic engineer in Tampere. “Using them, we built an algorithm that can predict the movement of vehicles and pedestrians on the street. The new solution has many potential uses in addition to boosting traffic safety. We already receive information about vehicle traffic, but not very much about pedestrian traffic. We want to see how people move, and perhaps construct heat maps of Tampere’s pedestrian flows to assist with traffic planning.”
Stenman added that a future possibility for the technology is to introduce more intelligence to traffic lights by identifying and predicting people flows.
“This implementation also provided one piece in the puzzle of autonomous vehicle systems,” said Torkkola. “A critical question is how self-driving vehicles are able to recognize and avoid obstacles. This type of pedestrian-recognition system could be an important element in the safety of autonomous vehicles in urban areas.”
In the UK, researchers have found that onboard, in-vehicle sensors could help reduce the effect of head traumas in the event of a crash. TRL is working with Imperial College of London as part of a multidisciplinary research collaboration that seeks to advance ways of diagnosing traumatic brain injuries (TBIs) caused during road-traffic collisions. Bringing together international experts from across the fields of vehicle safety, trauma biomechanics, neurosurgery, and emergency medicine, the aim of the AutoTriage project is to develop a novel technology that will benefit drivers and passengers.
“With the European Parliament’s recent historic adoption of groundbreaking updates to the General and Pedestrian Safety Regulation (GSR), a total of 17 life-saving advanced vehicle safety technologies will be introduced across the EU vehicle fleet,” said Dr. Phil Martin, Head of Biomechanics at TRL.
“Event data recorders (EDRs) are fundamental to the GSR requirements, which mandate the installation of EDRs on new vehicle types from 2022 onwards,” he added. “These regulations require the collection of anonymized data from the full suite of in-vehicle sensors, using these to support in-depth collision investigations and future vehicle safety research. The AutoTriage project, therefore, provides an important opportunity to maximize the future life-saving potential of EDRs fitted across the future EU fleet.”
The development team claims that its approach of using in-vehicle sensor data to perform real-time predictions of TBI severity and pathologies during road traffic collisions is a novel one. If successful and the full potential is realized, Martin believes that it would provide emergency responders with an immediate prediction of TBI risk and pathology.
“This would allow resources to be deployed through the emergency response chain in the most effective and timely manner possible. The potential life-saving benefits of AutoTriage cannot therefore be underestimated,” he concluded.