Abstract:Motion prediction modules are crucial for autonomous vehicles to forecast the future behavior of surrounding road users. Failures in prediction modules can mislead a downstream planner to make unsafe decisions. Currently, deep learning technology has been widely used to design prediction models due to its impressive performance. However, such models may fail in long-tail driving scenarios where the training data are insufficient or unavailable, which represents the so-called epistemic uncertainty of prediction models. This paper proposes a risk-aware decision-making (RADM) framework to handle the epistemic uncertainty arising from training the prediction model on insufficient data. First, a multi-agent prediction network with epistemic uncertainty quantification is proposed. This network uses the historical states of nearby road users, map information, and traffic lights as inputs. Then, the RADM utilizes model predictive control technique to not only process the multi-agent prediction results but also to consider the epistemic uncertainty of the prediction model. In addition, the accuracy of the established prediction model is verified on real-world driving datasets. Furthermore, the proposed RADM is evaluated on the log-replay data obtained from real-world driving logs and using the SUMO simulator, considering multiple challenging cases where pedestrians and non-motorized vehicles cross the intersection illegally. The experimental results demonstrate that RADM can reduce the driving risk and improve driving safety and supplementary videos are provided at https://github.com/SOTIF-AVLab/RADM.

Accident Prediction System Based on Hidden Markov Model for Vehicular Ad-Hoc Network in Urban Environments

A Method for Crash Prediction and Avoidance Using Hidden Markov Models

Real-Time Crash Risk Prediction using Long Short-Term Memory Recurrent Neural Network

Deep Learning-based Prediction of Traffic Accident Risk in Vehicular Networks

Vehicle Accident Prediction Based on Hidden Markov Model

A Novel Framework for Road Traffic Risk Assessment with HMM-Based Prediction Model.

A Deep Learning Approach to the Prediction of Short-term Traffic Accident Risk

Traffic Accident Severity Prediction Model using AI

Real-time Prediction of Accident using Big Data System

Research on Traffic Accident Prediction Model Based on Convolutional Neural Networks in VANET

A Data-Driven Approach for Driving Safety Risk Prediction Using Driver Behavior and Roadway Information Data

An urban traffic accident prediction model based on the full range of human, vehicle, road and environmental factors

Urban Traffic Accident Risk Prediction Revisited: Regionality, Proximity, Similarity and Sparsity

On Prediction Method of Traffic Risks with the Bayesian Network in Smart Transportation

A Probabilistic Prediction Model for the Safety Assessment of HDVs under Complex Driving Environments.

Trajectory Prediction for Human-driven Vehicles Based on a Combined Method of Deep Neural Network and Driving Risk Map

Accident Prediction Models for Urban Arterial System

Prediction Failure Risk-Aware Decision-Making for Autonomous Vehicles on Signalized Intersections

Vehicle Acceleration Prediction Based on Machine Learning Models and Driving Behavior Analysis

Development of a Safety Prediction Method for Arterial Roads Based on Big-Data Technology and Stacked AutoEncoder-Gated Recurrent Unit

A Scenario-Adaptive Driving Behavior Prediction Approach to Urban Autonomous Driving