CRAT-Pred: Vehicle Trajectory Prediction with Crystal Graph Convolutional Neural Networks and Multi-Head Self-Attention

Schmidt, Julian; Jordan, Julian; Gritschneder, Franz; Dietmayer, Klaus

doi:10.1109/icra46639.2022.9811637

Cited by 31 publications

(13 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HetEdgeGAT was selected for all further experiments because it uses the least number of parameters. Also, Schmidt et al [33] show that the resulting attention weights of interacts relations offer additional interpretability, as they are a direct measure for interactions.…”

Section: F Quantitative Resultsmentioning

confidence: 99%

SCENE: Reasoning About Traffic Scenes Using Heterogeneous Graph Neural Networks

Monninger

Schmidt

Rupprecht

et al. 2023

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

Understanding traffic scenes requires considering heterogeneous information about dynamic agents and the static infrastructure. In this work we propose SCENE, a methodology to encode diverse traffic scenes in heterogeneous graphs and to reason about these graphs using a heterogeneous Graph Neural Network encoder and task-specific decoders. The heterogeneous graphs, whose structures are defined by an ontology, consist of different nodes with type-specific node features and different relations with type-specific edge features. In order to exploit all the information given by these graphs, we propose to use cascaded layers of graph convolution. The result is an encoding of the scene. Task-specific decoders can be applied to predict desired attributes of the scene. Extensive evaluation on two diverse binary node classification tasks show the main strength of this methodology: despite being generic, it even manages to outperform task-specific baselines. The further application of our methodology to the task of node classification in various knowledge graphs shows its transferability to other domains.

show abstract

Section: F Quantitative Resultsmentioning

confidence: 99%

SCENE: Reasoning About Traffic Scenes Using Heterogeneous Graph Neural Networks

Monninger

Schmidt

Rupprecht

et al. 2023

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Referring to the spatial attention module in BAM [13], we first adopt a 1 × 1 convolution to compress the input feature X ′ , then adopt two 3 × 3 dilated convolutions to expand the received information, and lately, we adopt a 1 × 1 convolution to restore the spatial feature with dimension R H×W ×1 , and add a batch normalization layer at the end to adjust the output scale. The spatial dimension attention G w (X ′ ) is calculated as shown in Eq (6).…”

Section: Encoder-decoder Channel-spatial Convolutional Networkmentioning

confidence: 99%

“…In this section, the proposed CSTAE is compared with six baselines on comma2k19, KITTI, CCSAD datasets: (1)DD-LSTM [5], (2)CRATPred [6], (3)HCAGCN [7], (4)SABeRVAE [10], (5)MemMC-MAE [17], (6)SaMa-WCNN [18]. Table 3 shows the average P recision (mP re), Recall (mRec), and F 1 score (mF 1) after 10 runs for CSTAE model and its baselines on the dataset.…”

Section: Performance Comparisonmentioning

confidence: 99%

Sensor Spoofing Detection On Autonomous Vehicle Using Channel-spatial-temporal Attention Based Autoencoder Network

Zhou

Han

2023

Preprint

View full text Add to dashboard Cite

Autonomous vehicles rely on various sensors to evaluate the driving environment and issue essential control commands. Nonetheless, these sensors are susceptible to false data injection and spoofing attacks, which could easily be launched wirelessly and remotely by attackers. This paper proposes a channel-spatial-temporal attention-based autoencoder network to detect sensor spoofing attacks on autonomous vehicles. The network utilizes the reconstruction error based on the autoencoder to detect abnormalities in input time series data of multiple sensors. The proposed model consists of a memory-augmented based spatial-attention block and PSE-Res2Net block based encoder and decoder. PSE-Res2Net block initially adopts Res2Net module to generate multi-scale feature graph and enhance multi-dimensional representation ability of neural network, then applies the PSENet module to capture location-aware channel information and channel-sensitive spatial information through the interaction of channel attention and spatial attention. Moreover, the memory-augmented based temporal-attention block is developed to integrate multi-scale features and gather global sequence information of sensor measurements. The results of the experimental evaluation on the comma2k19, KITTI, and CCSAD datasets illustrate that the proposed detection model is superior to the baseline technologies in terms of mPre, mRec, mF1 score and achieves stronger robustness against noise.

show abstract

“…• CRAT-Pred [19] is a prediction model achieving stateof-the-art performance without requiring map information. Interactions are modeled with graph convolution and self-attention.…”

Section: Analysis Of Trajectory Prediction Modelsmentioning

confidence: 99%

MEAT: Maneuver Extraction from Agent Trajectories

Schmidt

Jordan

David

et al. 2022

2022 IEEE Intelligent Vehicles Symposium (IV)

Self Cite

View full text Add to dashboard Cite

Advances in learning-based trajectory prediction are enabled by large-scale datasets. However, in-depth analysis of such datasets is limited. Moreover, the evaluation of prediction models is limited to metrics averaged over all samples in the dataset. We propose an automated methodology that allows to extract maneuvers (e.g., left turn, lane change) from agent trajectories in such datasets. The methodology considers information about the agent dynamics and information about the lane segments the agent traveled along. Although it is possible to use the resulting maneuvers for training classification networks, we exemplary use them for extensive trajectory dataset analysis and maneuver-specific evaluation of multiple state-of-the-art trajectory prediction models. Additionally, an analysis of the datasets and an evaluation of the prediction models based on the agent dynamics is provided.

show abstract

CRAT-Pred: Vehicle Trajectory Prediction with Crystal Graph Convolutional Neural Networks and Multi-Head Self-Attention

Cited by 31 publications

References 38 publications

SCENE: Reasoning About Traffic Scenes Using Heterogeneous Graph Neural Networks

SCENE: Reasoning About Traffic Scenes Using Heterogeneous Graph Neural Networks

Sensor Spoofing Detection On Autonomous Vehicle Using Channel-spatial-temporal Attention Based Autoencoder Network

MEAT: Maneuver Extraction from Agent Trajectories

Contact Info

Product

Resources

About