Attention-based Neural Network for Driving Environment Complexity Perception

TL;DR

This paper introduces an attention-based neural network that accurately predicts the complexity of driving environments from videos and vehicle data, aiding autonomous vehicle perception systems.

Contribution

It presents a novel neural network model integrating attention mechanisms for environment complexity prediction using naturalistic driving data.

Findings

Achieved 91.22% classification accuracy on BDD dataset.

Effectively predicts environment complexity levels for autonomous vehicle perception.

Demonstrates the model's potential for enhancing AV safety and decision-making.

Abstract

Environment perception is crucial for autonomous vehicle (AV) safety. Most existing AV perception algorithms have not studied the surrounding environment complexity and failed to include the environment complexity parameter. This paper proposes a novel attention-based neural network model to predict the complexity level of the surrounding driving environment. The proposed model takes naturalistic driving videos and corresponding vehicle dynamics parameters as input. It consists of a Yolo-v3 object detection algorithm, a heat map generation algorithm, CNN-based feature extractors, and attention-based feature extractors for both video and time-series vehicle dynamics data inputs to extract features. The output from the proposed algorithm is a surrounding environment complexity parameter. The Berkeley DeepDrive dataset (BDD Dataset) and subjectively labeled surrounding environment complexity levels are used for model training and validation to evaluate the algorithm. The proposed attention-based network achieves 91.22% average classification accuracy to classify the surrounding environment complexity. It proves that the environment complexity level can be accurately predicted and applied for future AVs' environment perception studies.