Uncrewed Systems Technology 044 l Xer Technolgies X12 and X8 l Lidar sensors l Stan UGV l USVs insight l AUVSI Xponential 2022 l Cobra Aero A99H l Accession Class USV l Connectors I Oceanology International 2022

12 Researchers in the US have developed a way to predict the behaviour of pedestrians and cyclists around a driverless car (writes Nick Flaherty). The algorithm, developed at MIT, breaks down the problem of predicting the action of multiple users, each represented as an individual agent, into smaller pieces and tackles each one individually to solve the problem in parallel in real time. The behaviour prediction framework first guesses the relationships between two road users – which car, cyclist or pedestrian has the right of way, and which agent will yield – and uses those relationships to predict future trajectories for multiple agents. These estimated trajectories were found to be more accurate than those from other machine learning models, compared to real traffic flow in an enormous dataset compiled by autonomous driving company Waymo. The MIT technique even outperformed Waymo’s recently published model. Because the researchers broke the problem into simpler pieces, their technique used less memory. “This is a very intuitive idea, but no-one has fully explored it before, and it works quite well. The simplicity is definitely a plus,” said Xin Huang, a member of MIT’s Computer Science and Artificial Intelligence Laboratory. “We are comparing our model with other state-of-the-art models in the field, including the one from Waymo, and our model achieves top performance on this challenging benchmark. This has a lot of potential for the future.” The researchers’ machine learning method, called M2I, takes two inputs – past trajectories of the cars, cyclists and pedestrians interacting in a traffic setting such as a four-way intersection, and a map with street locations, lane configurations and so on. Using this information, a relation predictor infers which of two agents has the right of way, classifying one as a passer and the other as a yielder. Then a prediction model, known as a marginal predictor, guesses the trajectory for the passing agent, as this agent behaves independently. A second prediction model, known as a conditional predictor, then guesses what the yielding agent will do based on the actions of the passing agent. The system predicts a number of different trajectories for the yielder and passer, computes the probability of each one individually, and then selects the six joint results with the highest likelihood of occurring. M2I outputs a prediction of how these agents will move through traffic for the following 8 seconds. In one example, the MIT method caused a vehicle to slow down so that a pedestrian could cross the street, then sped up when they cleared the intersection. In another example, the vehicle waited until several cars had passed before turning from a side street onto a busy, main road. While this initial research focuses on interactions between two agents, M2I could infer relationships among many agents and then guess their trajectories by linking multiple marginal and conditional predictors. The researchers trained the models using the Waymo Open Motion Dataset, which contains millions of real traffic scenes involving vehicles, pedestrians and cyclists recorded by Lidar sensors and cameras mounted on the company’s autonomous vehicles. “Rather than just building a more complex model to solve this problem, we took an approach that is more like how a human thinks when they reason about interactions with others,” said Huang. “We don’t reason about hundreds of combinations of future behaviours, we make decisions quite fast.” However, the framework can’t account for cases where two agents are mutually influencing each other, such as when two vehicles each nudge forward at a four-way stop because the drivers aren’t sure who should be yielding. The researchers plan to address this limitation in future work. They also want to use their method to simulate realistic interactions between road users, which could be used to verify planning algorithms for self-driving cars or create huge amounts of synthetic driving data to improve model performance. Code spots possible trouble Driverless cars June/July 2022 | Uncrewed Systems Technology The MIT algorithm has improved the way driverless cars detect multiple pedestrians and cyclists