search

UMD     This Site





The authors predict the long-term (3-5 seconds) trajectories of road-agents, as well as their behavior (e.g. overspeeding, braking, etc.), in urban traffic scenes.

The authors predict the long-term (3-5 seconds) trajectories of road-agents, as well as their behavior (e.g. overspeeding, braking, etc.), in urban traffic scenes.

 

New work by the Geometric Algorithms for Modeling, Motion and Animation (GAMMA) research team uses a combination of spectral graph analysis and deep learning for traffic forecasting in urban traffic scenarios. The research will be of use to future autonomous vehicles. The team includes Rohan Chandra, Tianrui Guan, Srujan Panuganti, Trisha Mittal, Uttaran Bhattacharya, Aniket Bera, and ISR-affiliated Professor Dinesh Manocha (CS/ECE/UMIACS).

In their paper, Forecasting Trajectory and Behavior of Road agents Using Spectral Clustering in Graph-LSTMs, the researchers predict both low-level information (future trajectories) as well as high-level information (road agent behavior) from the extracted trajectory of each road agent. The formulation represents the proximity between the road agents using a dynamic weighted traffic-graph.

A two-stream graph convolutional LSTM network is used to perform traffic forecasting using these weighted traffic-graphs. The first stream predicts the spatial coordinates of road agents, while the second stream predicts whether a road agent is going to exhibit aggressive, conservative, or normal behavior. The researchers introduce spectral cluster regularization to reduce error margin in long-term prediction (3-5 seconds) and improve the accuracy of predicted trajectories. They evaluated their approach on the Argoverse, Lyft, and Apolloscape datasets and highlighted the benefits over prior trajectory prediction methods.

In practice, the new approach reduces average prediction error by more than 54% over prior algorithms and achieves a weighted average accuracy of 91.2% for behavior prediction.

The method has some limitations. Only one feature was used to design the behavior prediction model, which may not be able to generalize to new traffic scenarios. In addition, training is slow and takes several hours due to the number of computations required for computing the traffic graphs and corresponding Laplacian matrices. As the research progresses, behavior prediction will become data-driven, rather than rule-based. In addition, implementation will be optimized using GPU parallelization to improve the runtime.



Related Articles:
Who's walking deceptively? Manocha's team thinks they know.
AlphaGo family of AI programs grew from AMS simulation-based algorithms developed at UMD
Add-on mechanism could increase safety for networked vehicles
Realistic simulator improves safety of self-driving vehicles before road testing
Do Good Robotics Symposium to explore technologies that benefit society and the planet
Maryland research contributes to Google’s AlphaGo AI system
Clark School researchers test decentralized task allocation algorithms for UAV teams
ISR alum honors former advisor Benjamin Kedem with quantile frequency analysis paper
Manocha Keynote Speaker at 2019 ODSC Conference
ISR, ECE, CS, UMIACS faculty present 12 talks at Northrop Grumman University Research Symposium

December 9, 2019


«Previous Story  

 

 

Current Headlines

Ephremides and colleagues consider Age of Information scheduling problem for interconnected sensors

Clark School researchers test decentralized task allocation algorithms for UAV teams

Nuno Martins, alum Shinkyu Park and Jeff Shamma lead tutorial session at IEEE CDC 2019

UMD Online Engineering Graduate Program Ranks 18th Nationally

ISR alum honors former advisor Benjamin Kedem with quantile frequency analysis paper

Hybrid compositional planning for UAV rescue missions

Who's walking deceptively? Manocha's team thinks they know.

Chakraborty, Seneviratne, and Zou Named Outstanding Graduate Assistants

Data fusion modeling can estimate residential radon levels

Ryzhov, Chen develop theoretical framework for approximate Bayesian learning

 
 
Back to top  
Home Clark School Home UMD Home