Research out of University of Leeds could help self-driving cars become more human-friendly. By investigating how to better understand human behavior in traffic, neuroscientific theories of how the brain makes decisions could enable automated vehicle technology to predict when pedestrians are going to cross the road.
Drift Diffusion Model
The decision-making model explored by the team of researchers is called drift diffusion, and it could be used in scenarios involving a car giving way to a pedestrian, with or without signals. Through this prediction capability, the autonomous vehicle could communicate more effectively with pedestrians. It would achieve a better understanding of their movements in traffic and external signals like flashing lights, which would help maximise traffic flow and decrease uncertainty.
Drift diffusion models rely on the assumption that people reach decisions after they accumulate sensory evidence up to a threshold, at which point the decision is made.
Professor Gustav Markkula is from the University of Leeds’ Institute for Transport Studies. He is the lead author of the study.
“When making the decision to cross, pedestrians seem to be adding up lots of different sources of evidence, not only relating to the vehicle’s distance and speed, but also using communicative cues from the vehicle in terms of deceleration and headlight flashes,” Professor Markkula said.
“When a vehicle is giving way, pedestrians will often feel quite uncertain about whether the car is actually yielding, and will often end up waiting until the car has almost come to a full stop before starting to cross,” he continued. “Our model clearly shows this state of uncertainty borne out, meaning it can be used to help design how automated vehicles behave around pedestrians in order to limit uncertainty, which can improve both traffic safety and traffic flow.”
“It is exciting to see that these theories from cognitive neuroscience can be brought into this type of real-world context and find an applied use.”
Testing the Model
The team set out to test the model with virtual reality. Trial participants were placed in different road-crossing scenarios within the university’s HIKER (Highly Immersive Kinematic Experimental Research) pedestrian simulator. Their movements were tracked while walking freely inside a stereoscopic 3D virtual scene that presented oncoming traffic. The participants were told to cross the road when they felt safe enough.
The researchers tested multiple different scenarios, including the approaching vehicle maintaining a constant speed and decelerating to let the pedestrian cross. The vehicle also sometimes flashed its headlights to signal a cross.
The tests demonstrated that the participants seemingly added up the sensory data from vehicle distance, speed, acceleration, and communicative cues before making a decision on when to cross. This indicated to the team that the drift diffusion model could predict if, and when, pedestrians would likely cross the road.
“These findings can help provide a better understanding of human behavior in traffic, which is needed both to improve traffic safety and to develop automated vehicles that can coexist with human road users,” Professor Markulla said.
“Safe and human-acceptable interaction with pedestrians is a major challenge for developers of automated vehicles, and a better understanding of how pedestrians behave will be key to enable this.”
According to lead author Dr. Jami Pekkanen, “Predicting pedestrian decisions and uncertainty can be used to optimise when, and how, the vehicle should decelerate and signal to communicate that it's safe to cross, saving time and effort for both.”
- Quantum-Enhanced AI Revolutionizes Cancer Drug Discovery: A Leap Forward with Industrial Generative AI
- Exploring Gemini 1.5: How Google’s Latest Multimodal AI Model Elevates the AI Landscape Beyond Its Predecessor
- Danny Postma, Founder of HeadshotPro – Interview Series
- ChatGPT Team: Introducing OpenAI’s Team Productivity Booster
- Using AI-Mechanized Hyperautomation for Organizational Decision Making