Predicting Others Behavior on the Road With Artificial Intelligence – SciTechDaily
Researchers have created a machine-learning system that efficiently predicts the future trajectories of multiple road users, like drivers, cyclists, and pedestrians, which could enable an autonomous vehicle to more safely navigate city streets. If a robot is going to navigate a vehicle safely through downtown Boston, it must be able to predict what nearby drivers, cyclists, and pedestrians are going to do next. Credit: MIT
A new machine-learning system may someday help driverless cars predict the next moves of nearby drivers, pedestrians, and cyclists in real-time.
Humans may be one of the biggest roadblocks to fully autonomous vehicles operating on city streets.
If a robot is going to navigate a vehicle safely through downtown Boston, it must be able to predict what nearby drivers, pedestrians, and cyclists are going to do next.
Behavior prediction is a tough problem, however, and current artificial intelligence solutions are either too simplistic (they may assume pedestrians always walk in a straight line), too conservative (to avoid pedestrians, the robot just leaves the car in park), or can only forecast the next moves of one agent (roads typically carry many users at once.)
MIT researchers have devised a deceptively simple solution to this complicated challenge. They break a multiagent behavior prediction problem into smaller pieces and tackle each one individually, so a computer can solve this complex task in real-time.
These simulations show how the system the researchers developed can predict the future trajectories (shown using red lines) of the blue vehicles in complex traffic situations involving other cars, bicyclists, and pedestrians. Credit: MIT
Their behavior-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 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 Waymos recently published model. And 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. We are comparing our model with other state-of-the-art models in the field, including the one from Waymo, the leading company in this area, and our model achieves top performance on this challenging benchmark. This has a lot of potential for the future, says co-lead author Xin Cyrus Huang, a graduate student in the Department of Aeronautics and Astronautics and a research assistant in the lab of Brian Williams, professor of aeronautics and astronautics and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL).
Joining Huang and Williams on the paper are three researchers from Tsinghua University in China: co-lead author Qiao Sun, a research assistant; Junru Gu, a graduate student; and senior author Hang Zhao PhD 19, an assistant professor. The research will be presented at the Conference on Computer Vision and Pattern Recognition.
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, etc.
Using this information, a relation predictor infers which of two agents has the right of way first, classifying one as a passer and one as a yielder. Then a prediction model, known as a marginal predictor, guesses the trajectory for the passing agent, since 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 next eight seconds. In one example, their method caused a vehicle to slow down so a pedestrian could cross the street, then speed 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 (light detection and ranging) sensors and cameras mounted on the companys autonomous vehicles. They focused specifically on cases with multiple agents.
To determine accuracy, they compared each methods six prediction samples, weighted by their confidence levels, to the actual trajectories followed by the cars, cyclists, and pedestrians in a scene. Their method was the most accurate. It also outperformed the baseline models on a metric known as overlap rate; if two trajectories overlap, that indicates a collision. M2I had the lowest overlap rate.
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. A human does not reason about all hundreds of combinations of future behaviors. We make decisions quite fast, Huang says.
Another advantage of M2I is that, because it breaks the problem down into smaller pieces, it is easier for a user to understand the models decision-making. In the long run, that could help users put more trust in autonomous vehicles, says Huang.
But the framework cant account for cases where two agents are mutually influencing each other, like when two vehicles each nudge forward at a four-way stop because the drivers arent sure who should be yielding.
They 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.
Predicting future trajectories of multiple, interacting agents is under-explored and extremely challenging for enabling full autonomy in complex scenes. M2I provides a highly promising prediction method with the relation predictor to discriminate agents predicted marginally or conditionally which significantly simplifies the problem, wrote Masayoshi Tomizuka, the Cheryl and John Neerhout, Jr. Distinguished Professor of Mechanical Engineering at University of California at Berkeley and Wei Zhan, an assistant professional researcher, in an email. The prediction model can capture the inherent relation and interactions of the agents to achieve the state-of-the-art performance. The two colleagues were not involved in the research.
Reference: M2I: From Factored Marginal Trajectory Prediction to Interactive Prediction by Qiao Sun, Xin Huang, Junru Gu, Brian C. Williams and Hang Zhao. 28 March 2022, Computer Science > Robotics.arXiv:2202.11884
This research is supported, in part, by the Qualcomm Innovation Fellowship. Toyota Research Institute also provided funds to support this work.
Read the original post:
Predicting Others Behavior on the Road With Artificial Intelligence - SciTechDaily
- Impact of Artificial Intelligence (AI) on Media and Creative Industries - EDMO - March 1st, 2025 [March 1st, 2025]
- 1 Artificial Intelligence (AI) Stock That Could Be Bigger Than Nvidia in 5 Years - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- This Artificial Intelligence (AI) Stock Is Up 15% in 2025 Already. It Is Still a Solid Buy? - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- Can artificial intelligence be the future solution to the enormous challenges and suffering caused by Schizophrenia? - Nature.com - March 1st, 2025 [March 1st, 2025]
- Applications of Artificial Intelligence in Medical Education: A Systematic Review - Cureus - March 1st, 2025 [March 1st, 2025]
- This Artificial Intelligence (AI) Stock Is Up 15% in 2025 Already. It Is Still a Solid Buy? - AOL - March 1st, 2025 [March 1st, 2025]
- Federal Executive Forum Artificial Intelligence Strategies in Government Progress and Best Practices 2025 - Federal News Network - March 1st, 2025 [March 1st, 2025]
- Introduction to Artificial Intelligence for General Surgeons: A Narrative Review - Cureus - March 1st, 2025 [March 1st, 2025]
- How is Artificial Intelligence Affecting Health Care? - Workers Comp Forum - March 1st, 2025 [March 1st, 2025]
- 1 Spectacular Artificial Intelligence (AI) Stock to Buy With $50 Right Now - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- 20+ Advantages and Disadvantages of AI | Pros of Artificial Intelligence - Simplilearn - March 1st, 2025 [March 1st, 2025]
- Prediction: This Top Artificial Intelligence (AI) Stock Will Start Skyrocketing After March 6 - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- 1 Surprising Stock Harnessing the Power of Artificial Intelligence (AI) - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- 2 Top Artificial Intelligence (AI) Stocks to Buy On the Dip Amid Nasdaq Selloff - Yahoo Finance - March 1st, 2025 [March 1st, 2025]
- Review: Artificial intelligence is shaping the future of diabetes care - News-Medical.Net - March 1st, 2025 [March 1st, 2025]
- Prediction: This Artificial Intelligence (AI) Stock -- a 1,020% Gainer Since Its IPO -- Won't Split Its Stock in 2025. Here's Why - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- A Nobel laureate on the economics of artificial intelligence - MIT Technology Review - March 1st, 2025 [March 1st, 2025]
- Prediction: This Top Artificial Intelligence (AI) Stock Will Start Skyrocketing After March 6 - Nasdaq - March 1st, 2025 [March 1st, 2025]
- Meta Platforms Just Caused This Crucial Artificial Intelligence (AI) Stock to Plummet. Should You Buy the Dip? - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- Artificial Intelligence - AI Update, February 28, 2025: AI News and Views From the Past Week - MarketingProfs.com - March 1st, 2025 [March 1st, 2025]
- The Ultimate Artificial Intelligence (AI) ETF to Buy With $50 Right Now - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- Prediction: This Artificial Intelligence (AI) Company Will Split Its Stock in 2025 - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- Should You Forget Nvidia and Buy 2 Artificial Intelligence (AI) Stocks Instead? - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- Why Artificial Intelligence Stocks SoundHound AI, IonQ, and C3.ai Are Struggling Today - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- 2 Top Artificial Intelligence (AI) Stocks Ready for a Bull Run - The Motley Fool - March 1st, 2025 [March 1st, 2025]
- AI Cant Do This Anymore: The Dangers of Artificial Intelligence in Academia - Skidmore News - March 1st, 2025 [March 1st, 2025]
- Whats Next in Artificial Intelligence: Agents that can do more than chatbots - Pittsburgh Post-Gazette - February 9th, 2025 [February 9th, 2025]
- Geopolitics of artificial intelligence to be focus of major summit in Paris; AP explains - Yahoo - February 9th, 2025 [February 9th, 2025]
- Geopolitics of artificial intelligence to be focus of major summit in Paris; AP explains - The Associated Press - February 9th, 2025 [February 9th, 2025]
- 3 Top Artificial Intelligence Stocks to Buy in February - MSN - February 9th, 2025 [February 9th, 2025]
- Geopolitics of artificial intelligence to be focus of major summit in Paris; AP explains - Lufkin Daily News - February 9th, 2025 [February 9th, 2025]
- 2 of the Hottest Artificial Intelligence (AI) Stocks on the Planet Can Plunge Up to 94%, According to Select Wall Street Analysts - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- These 2 Stocks Are Leading the Data Center Artificial Intelligence (AI) Trend, but Are They Buys Right Now? - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Book Review | Genesis: Artificial Intelligence, Hope, and the Human Spirit - LSE - February 9th, 2025 [February 9th, 2025]
- The Artificial Intelligence Action Summit In France: Maintaining The Dialogue On Global AI Regulation - Forrester - February 9th, 2025 [February 9th, 2025]
- Is prediction the next frontier for artificial intelligence? - Healthcare IT News - February 9th, 2025 [February 9th, 2025]
- The Artificial Intelligence in Medicines Market Is Set to Reach $18,119 Million | CAGR of 49.6% - openPR - February 9th, 2025 [February 9th, 2025]
- Geopolitics of artificial intelligence to be focus of major summit in Paris; AP explains - The Audubon County Advocate Journal - February 9th, 2025 [February 9th, 2025]
- Around and About with Richard McCarthy: Asking AI about itself: Will artificial intelligence ever surpass humankind? - GazetteNET - February 9th, 2025 [February 9th, 2025]
- Will the Paris artificial intelligence summit set a unified approach to AI governanceor just be another conference? - Bulletin of the Atomic... - February 9th, 2025 [February 9th, 2025]
- Apple Stock Jumps on Artificial Intelligence (AI) Driving iPhone Sales. Here's Why It's Not Getting Crushed by the DeepSeek Launch. - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Who will win the race to Artificial General Intelligence? - The Indian Express - February 9th, 2025 [February 9th, 2025]
- Prediction: This Artificial Intelligence (AI) Chip Stock Will Win Big From DeepSeek's Feat - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Prediction: 2 Artificial Intelligence (AI) Stocks That Will Be Worth More Than Nvidia 3 Years From Now - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- State of Louisiana Launches Innovation Brand, Announces Creation of $50 Million Growth Fund and Artificial Intelligence Research Institute - Louisiana... - February 9th, 2025 [February 9th, 2025]
- Using smart technologies and artificial intelligence in food packaging can reduce food waste - Yahoo News Canada - February 9th, 2025 [February 9th, 2025]
- BigBear.ai Wins Department of Defense Contract to Prototype Near-Peer Adversary Geopolitical Risk Analysis for Chief Digital and Artificial... - February 9th, 2025 [February 9th, 2025]
- Should Investors Change Their Artificial Intelligence (AI) Investment Strategy After the DeepSeek Launch? - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- 1 Unstoppable Artificial Intelligence (AI) Stock to Buy Before It Punches Its Ticket to the $4 Trillion Club - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Got 10 Years and $1000? These 3 Artificial Intelligence (AI) Stocks Are Set to Soar. - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- 1 Artificial Intelligence (AI) Stock Down 33% to Buy Hand Over Fist, According to Wall Street - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Rihanna Calls Out Use of Artificial Intelligence on Her Voice to Doctor a Clip of Her Speaking - Billboard - February 9th, 2025 [February 9th, 2025]
- 3 Best Artificial Intelligence (AI) Stocks to Buy in February - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Buying This Top Artificial Intelligence (AI) Stock Looks Like a No-Brainer Right Now - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Is Arm Stock a Buy After the Artificial Intelligence (AI) Chip Designer Released Its Quarterly Earnings Report? - The Motley Fool - February 9th, 2025 [February 9th, 2025]
- Artificial Intelligence, the Academy, And A New Studia Humanitatis - Minding The Campus - February 9th, 2025 [February 9th, 2025]
- The Trump Administrations Artificial Intelligence Rollback Is a Chance to Rethink AI Policy - Ms. Magazine - February 5th, 2025 [February 5th, 2025]
- Workday layoffs: California-based company lays off 1,750 employees, 8.5% of its workforce in favor of artificial intelligence - ABC7 Los Angeles - February 5th, 2025 [February 5th, 2025]
- It can really transform lives: Navigating the ethical landscape of artificial intelligence - WKMG News 6 & ClickOrlando - February 5th, 2025 [February 5th, 2025]
- Legal Restrictions Governing Artificial Intelligence in the Workplace - Law.com - February 5th, 2025 [February 5th, 2025]
- Google drops AI weapons banwhat it means for the future of artificial intelligence - VentureBeat - February 5th, 2025 [February 5th, 2025]
- MPs to scrutinise use of artificial intelligence in the finance sector - ComputerWeekly.com - February 5th, 2025 [February 5th, 2025]
- Catalyzing Change: Innovation and Efficiency through Artificial Intelligence in Contracting - United States Army - February 5th, 2025 [February 5th, 2025]
- STSD to hear cost breakdown, address artificial intelligence in education - The Wellsboro Gazette - February 5th, 2025 [February 5th, 2025]
- OECD activities during the Artificial Intelligence (AI) Action Summit - OECD - February 5th, 2025 [February 5th, 2025]
- Tether Ventures Into Artificial Intelligence With New Application Suite - Bitcoin.com News - February 5th, 2025 [February 5th, 2025]
- Will Artificial Intelligence Kill Acting? Nicholas Cage Thinks It Could - Movieguide - February 5th, 2025 [February 5th, 2025]
- 3 Reasons to Buy This Artificial Intelligence (AI) Stock on the Dip - The Motley Fool - February 5th, 2025 [February 5th, 2025]
- 1 No-Brainer Artificial Intelligence (AI) Stock to Buy With $35 and Hold for the Long Run - The Motley Fool - February 5th, 2025 [February 5th, 2025]
- Google renounces its promise not to develop weapons with artificial intelligence - Mezha.Media - February 5th, 2025 [February 5th, 2025]
- DeepSeek Just Changed Generative Artificial Intelligence (AI) Forever. 2 Surprising Winners From Its Innovation. - The Motley Fool - February 5th, 2025 [February 5th, 2025]
- FUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare - The BMJ - February 5th, 2025 [February 5th, 2025]
- DeepSeek Just Exposed the Biggest Flaw of the Artificial Intelligence (AI) Revolution - The Motley Fool - February 5th, 2025 [February 5th, 2025]
- Artificial Intelligence Is Here: How The Innovative Technology Is Taking Over The Stateline - WREX.com - February 5th, 2025 [February 5th, 2025]
- The Ultimate Artificial Intelligence (AI) Stocks to Buy in 2025 - The Motley Fool - February 5th, 2025 [February 5th, 2025]
- This Magnificent Artificial Intelligence (AI) Stock Has Shot Up Over 175% in Just 3 Months, and It Could Soar Higher in 2025 - The Motley Fool - February 5th, 2025 [February 5th, 2025]
- Artificial intelligence is bringing nuclear power back from the dead maybe even in California - CalMatters - February 5th, 2025 [February 5th, 2025]
- Got $5,000? These Are 3 of the Cheapest Artificial Intelligence Stocks to Buy Right Now - The Motley Fool - February 5th, 2025 [February 5th, 2025]
- Compass Capital partners with MIT Sloan School of Management on an artificial intelligence project - ZAWYA - February 5th, 2025 [February 5th, 2025]
- 3 No-Brainer Artificial Intelligence (AI) Stocks to Buy With $500 Right Now - The Motley Fool - February 5th, 2025 [February 5th, 2025]