Transformer based trajectory prediction

Postnikov, Aleksey; Gamayunov, Aleksander; Ferrer, Gonzalo

doi:10.48550/arxiv.2112.04350

Cited by 4 publications

(12 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SNGP is directly applied to the scene encoder and the uncertainty is measured by the predicted Gaussian process variance. Similarly, our uncertainty module is applied to the latent features of the scene encoder, but we use error regression to estimate the uncertainty similar to [11]. Our error regression network is trained to approximate the trajectory prediction error.…”

Section: Related Workmentioning

confidence: 99%

“…To accurately detect prediction errors in ID scenarios, we introduce an uncertainty estimation network, denoted as g u , which is a small MLP applied to the scene encoder. We formulate the problem of uncertainty estimation as a regression task, similar to [11], and train the neural network to predict the true trajectory prediction error e i for the agent i, given the encoder feature vector h i . During inference, the error regression network, we denote as E reg throughout the experiments, outputs the uncertainty as êi = g u (h i ).…”

Section: E Uncertainty Estimationmentioning

confidence: 99%

“…However, ensembles involve multiple forward passes and therefore have limited applicability in real-time systems like automated vehicles. VNT [10] combines a graph-based encoder on a vectorized scene with a transformer-based decoder and ViT [11] a vision transformerbased encoder on a rasterized scene with an MLP decoder. Both models predict a Gaussian mixture distribution over future trajectories, but only predict a deterministic mean without a covariance, which is set to the identity matrix Σ = I during training.…”

Section: B Comparison With State-of-the-artmentioning

confidence: 99%

“…The prior work assesses the reliability of trajectory prediction with OOD detection [8], [9] and uncertainty estimation [10], [11], [12]. OOD detection distinguishes between ID and OOD scenarios, while uncertainty estimation determines the certainty of the prediction regardless of whether it is ID or OOD.…”

Section: Introductionmentioning

confidence: 99%

“…A Gaussian mixture model on the latent representation learns the probability density function of the ID features during training and detects OOD samples in regions of the feature space with low likelihood of this density function. Secondly, we add an error regression network [11] on the scene encoder for uncertainty estimation and train it with a supervised regression loss to estimate the prediction error. During inference, the estimated error is used as uncertainty.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Benchmark for Unsupervised Anomaly Detection in Multi-Agent Trajectories

Wiederer

Schmidt

Kreßel

et al. 2022

2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

Despite the significant research efforts on trajectory prediction for automated driving, limited work exists on assessing the prediction reliability. To address this limitation we propose an approach that covers two sources of error, namely novel situations with out-of-distribution (OOD) detection and the complexity in in-distribution (ID) situations with uncertainty estimation. We introduce two modules next to an encoderdecoder network for trajectory prediction. Firstly, a Gaussian mixture model learns the probability density function of the ID encoder features during training, and then it is used to detect the OOD samples in regions of the feature space with low likelihood. Secondly, an error regression network is applied to the encoder, which learns to estimate the trajectory prediction error in supervised training. During inference, the estimated prediction error is used as the uncertainty. In our experiments, the combination of both modules outperforms the prior work in OOD detection and uncertainty estimation, on the Shifts robust trajectory prediction dataset by 2.8 % and 10.1 %, respectively. The code is publicly available 4 .

show abstract

Section: Related Workmentioning

confidence: 99%

Section: E Uncertainty Estimationmentioning

confidence: 99%

Section: B Comparison With State-of-the-artmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Benchmark for Unsupervised Anomaly Detection in Multi-Agent Trajectories

Wiederer

Schmidt

Kreßel

et al. 2022

2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

show abstract

Multi‐agent trajectory prediction with adaptive perception‐guided transformers

Nguyen,

Yoo

2024

IET Intelligent Trans Sys

View full text Add to dashboard Cite

The ability to predict the trajectory of an autonomous vehicle accurately is crucial for safe and efficient navigation. However, predicting diverse and multimodal futures can be challenging. Recent approaches such as attention and graph neural networks have achieved state‐of‐the‐art performance by considering agent interactions and map contexts. This study focused on multi‐agent prediction using an agent‐centric approach with transformers. This enables parallel computation and a comprehensive understanding of the environment. Two main features are introduced: an adaptive receptive field (ARF) that captures the relevant surroundings for each agent, and perception encoding, which serves as spatial context embeddings. The ARF adapts to the agent's velocity and rotation, focusing attention ahead at high speeds or to the sides when it is slower. Perception encoding divides agents or lanes into levels and encodes the information of each level. This approach enables the efficient encoding of complex spatial relationships. The proposed method combines these advances with transformer modelling for multi‐agent trajectory prediction while ensuring real‐time prediction capabilities. The approach is evaluated on the Argoverse benchmark and better performance than the state‐of‐the‐art baseline is achieved. By addressing challenges such as multimodal outputs and robustness, the study enhances the safety and efficiency of autonomous driving systems by more accurately predicting trajectories.

show abstract

Sea Drift Trajectory Prediction Based on Quantum Convolutional Long Short-Term Memory Model

Yan,

Zhang,

Parvej

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

This paper proposes a novel Sea Drift Trajectory Prediction method based on the Quantum Convolutional Long Short-Term Memory (QCNN-LSTM) model. Accurately predicting sea drift trajectories is a challenging task, as they are influenced by various complex factors, such as ocean currents, wind speed, and wave morphology. Therefore, in a complex marine environment, there is a need for more applicable and computationally advanced prediction methods. Our approach combines quantized convolutional neural networks with Long Short-Term Memory networks, utilizing two different input types of prediction to enhance the network’s applicability. By incorporating quantization techniques, we improve the computational power and accuracy of the trajectory prediction. We evaluate our method using sea drift datasets and AUV drift trajectory datasets, comparing it with other commonly used traditional methods. The experimental results demonstrate significant improvements in accuracy and robustness achieved by our proposed Quantum Convolutional Long Short-Term Memory model. Regardless of the input mode employed, the accuracy consistently surpasses 98%. In conclusion, our research provides a new approach for sea drift trajectory prediction, enhancing prediction accuracy and providing valuable insights for marine environmental management and related decision-making. Future research can further explore and optimize this model to have a greater impact on marine prediction and applications.

show abstract

Transformer based trajectory prediction

Cited by 4 publications

References 0 publications

A Benchmark for Unsupervised Anomaly Detection in Multi-Agent Trajectories

A Benchmark for Unsupervised Anomaly Detection in Multi-Agent Trajectories

Multi‐agent trajectory prediction with adaptive perception‐guided transformers

Sea Drift Trajectory Prediction Based on Quantum Convolutional Long Short-Term Memory Model

Contact Info

Product

Resources

About