\n<\/p>\n
Waymo has created a new computer model designed to more accurately answer a fundamental question: how does its autonomous driving software stack up against humans?<\/p>\n
The Alphabet-owned robotaxi company, which developed the computer model of human driving capabilities in conjunction with TU Delft, published a research paper<\/a> about it in Nature Communications on Wednesday. <\/p>\n Waymo said it expects the new model to be more accurate than the previous version it has used over the past several years. The new model was built using a framework called active inference \u2014 the theory that a driver is constantly imagining possible futures and taking actions to reach the safest, most predictable one. <\/p>\n Waymo said the new model will help it better understand how humans behave in crash scenarios that its robotaxis encounter. <\/p>\n \u201cFor decades, the automotive industry has used physical and virtual crash dummies to evaluate a car\u2019s safety features, including its hardware and structural integrity,\u201d Waymo wrote in a blog post on Wednesday. The new model, Waymo said, \u201cevolves this concept, serving as a behavioral benchmark for autonomous driving systems able to realistically represent reasonable expectations on how a careful and competent human driver responds to traffic conflicts.\u201d<\/p>\n A more accurate model of human driving behavior is table stakes for autonomous vehicle companies that need to understand and grade the performance of its robotaxis in crashes. And it comes at a critical juncture for Waymo, which is scaling to more cities and facing greater scrutiny from regulators and the public. <\/p>\n In January, when a Waymo robotaxi struck a child<\/a> near a school in Santa Monica, California, the company relied on its previous computer model to claim that an attentive human driver would have made impact at around 14 miles per hour. The Waymo robotaxi hit the child at just 6 miles per hour, after decelerating from 17 miles per hour, and the company said she sustained minor injuries. (The crash is still under investigation by the National Highway Traffic Safety Administration and the National Transportation Safety Board.)<\/p>\n The biggest difference between this new model \u2014 which Waymo calls the Reference Driver \u2014 and its predecessor is that it is able to reproduce a human driver\u2019s behavior in the run-up to a crash. Previously, Waymo\u2019s models (and other industry models) focused on replicating \u201clast-second, reactive\u201d human maneuvers, according to the company.<\/p>\n The Reference Driver, meanwhile, can \u201csimulate the internal \u2018surprise\u2019 a driver feels during a conflict, providing a more human-like benchmark for autonomous driving systems that was previously impossible to automate at scale,\u201d Arkady Zgonnikov, an assistant professor at TU Delft, said in a statement.<\/p>\n Waymo says this new driver model can be adapted to model a \u201cwide range of road user behaviors beyond collision avoidance,\u201d and that it is better-equipped to be applied to \u201clarge test sets with thousands of scenarios.\u201d <\/p>\n \u201cThe model can represent and evaluate numerous complex, real-world crashes in a virtual environment, identifying performance improvements with unprecedented speed and efficiency,\u201d the company wrote.<\/p>\n Waymo wants others to collaborate on pushing the Reference Driver further, too. The company said Wednesday that it is making the research code for the model available under an academic, non-commercial license that allows it to be used for research, teaching, personal experimentation, and scientific publication. <\/p>\n<\/div>\n When you purchase through links in our articles, we may earn a small commission<\/a>. This doesn\u2019t affect our editorial independence.<\/em><\/p>\n