HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling

Huang, Xin; Rosman, Guy; Gilitschenski, Igor; Jasour, Ashkan; McGill, Stephen G.; Leonard, John J.; Williams, Brian

doi:10.1109/icra46639.2022.9812254

Cited by 16 publications

(6 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, several techniques to predict future trajectories of surrounding vehicles, based on continuous/discrete hybrid models, e.g., based on Hidden Markov Models [32] or Long Short-Term Memory (LSTM) models [33,34] have been proposed in the literature. However, it is very difficult to derive practically useful and statistically valid concentration bounds on the obtained probability distributions for the purpose of constructing ambiguity sets.…”

Section: Related Workmentioning

confidence: 99%

Safe, learning-based MPC for highway driving under lane-change uncertainty: A distributionally robust approach

Schuurmans

Katriniok

Meissen

et al. 2023

Artificial Intelligence

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Safe, learning-based MPC for highway driving under lane-change uncertainty: A distributionally robust approach

Schuurmans

Katriniok

Meissen

et al. 2023

Artificial Intelligence

View full text Add to dashboard Cite

“…As MAP metric is typical for detection tasks we decided to use a typical approach -non-maximum suppression. This algorithm has been successfully used in multiple of previous works [5,7,12,14,18]. More specifically, in case where two trajectories appear to be close enough to each other the less probable was suppressed in favour for more probable one.…”

Section: Postprocessingmentioning

confidence: 99%

MPA: MultiPath++ Based Architecture for Motion Prediction

Konev¹

2022

Preprint

View full text Add to dashboard Cite

Autonomous driving technology is developing rapidly and nowadays first autonomous rides are being provided in city areas. This requires the highest standards for the safety and reliability of the technology. Motion prediction part of the general self-driving pipeline plays a crucial role in providing these qualities. In this work we present one of the solutions for Waymo Motion Prediction Challenge 2022 based on MultiPath++ [16] ranked the 3rd as of May, 26 2022. Our source code is publicly available on GitHub 1 .1 https : / / github . com / stepankonev / waymo -motionprediction-chalenge-2022-multipath-plus-plus

show abstract

“…Predictions of deterministic trajectories involve methods where forthcoming paths are determined based on their initial parameter values. Deterministic methodologies can be designed using statistical [15], artificial intelligence [16], and hybrid [17] models. Receding Horizon techniques that utilize kinematic and kinetic models have been the dominant type applied and have demonstrated their effectiveness [18,19].…”

Section: Introductionmentioning

confidence: 99%

Aircraft Trajectory Prediction Enhanced through Resilient Generative Adversarial Networks Secured by Blockchain: Application to UAS-S4 Ehécatl

Hashemi,

Botez

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

This paper introduces a novel and robust data-driven algorithm designed for Aircraft Trajectory Prediction (ATP). The approach employs a Neural Network architecture to predict future aircraft trajectories, utilizing input variables such as latitude, longitude, altitude, heading, speed, and time. The model’s foundation is rooted in the Generative Adversarial Network (GAN) framework, known for its inherent generative capabilities, rendering it remarkably resilient against Adversarial Attacks. To enhance its credibility, the Blockchain is employed as a Ledger Technology (LT) to securely store legitimate predicted values utilized in subsequent trajectory predictions. The Blockchain ensures that only authorized and non-adversarial samples are stored in the blocks, rejecting any adversarial predictions. In the validation process, trajectory data for training the GAN model were generated through the UAS-S4 Ehécatl simulation model. The performance evaluation relies on the model’s resistance to adversarial attacks, measured by fooling rates. The results acquired affirm the excellent efficacy of the GAN model, Secured by Blockchain, approaching against adversarial attacks.

show abstract

HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling

Cited by 16 publications

References 32 publications

Safe, learning-based MPC for highway driving under lane-change uncertainty: A distributionally robust approach

Safe, learning-based MPC for highway driving under lane-change uncertainty: A distributionally robust approach

MPA: MultiPath++ Based Architecture for Motion Prediction

Aircraft Trajectory Prediction Enhanced through Resilient Generative Adversarial Networks Secured by Blockchain: Application to UAS-S4 Ehécatl

Contact Info

Product

Resources

About