Cyclist Trajectory Prediction Using Bidirectional Recurrent Neural Networks

Saleh, Khaled J.; Hossny, Mohammed; Nahavandi, Saeid

doi:10.1007/978-3-030-03991-2_28

Cited by 17 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bisagno et al (2018) added group coherence information in the social pooling layer. Saleh et al predicted trajectories of pedestrians (Saleh et al, 2018b) and cyclists (Saleh et al, 2018a), adapting the LSTM architecture for the perspective of a moving vehicle. Numerous other implementations of the LSTM-based predictors offer various improvements, such as increased generalizability to new and crowded environments (Shi et al, 2019; Xue et al, 2019), considering the immediate (Zhang et al, 2019) or long-term (Xue et al, 2017) intention of the agents, augmenting the state of the person with the head pose (Hasan et al, 2018) or adding a better pooling mechanism with relative importance of each person in the vicinity of the target agent (Fernando et al, 2018; Pei et al, 2019; Xu et al, 2018).…”

Section: Pattern-based Approachesmentioning

confidence: 99%

“…Many pattern-based (Bennewitz et al, 2005, 2002; Carvalho et al, 2019; Chen et al, 2016, 2008; Goldhammer et al, 2014; Habibi et al, 2018; Han et al, 2019; Hermes et al, 2009; Huynh and Alaghband, 2019; Kim et al, 2017, 2011; Kucner et al, 2017, 2013; Makansi et al, 2019; Makris and Ellis, 2002; Molina et al, 2018; Nikhil and Tran Morris, 2018; Piciarelli et al, 2005; Ridel et al, 2019; Saleh et al, 2018b; Sung et al, 2012; Suraj et al, 2018; Tadokoro et al, 1993; Thompson et al, 2009; Unhelkar et al, 2015; Wang et al, 2016; Xiao et al, 2015; Xue et al, 2019, 2017) and planning-based methods (Gong et al, 2011; Karasev et al, 2016; Kitani et al, 2012; Rhinehart et al, 2018a; Rudenko et al, 2017; Vasquez, 2016; Yen et al, 2008; Ziebart et al, 2009) are unaware predictors, owing to the increase of complexity for conditioning the learned transition patterns or optimal actions on the presence and positions of other agents. Methods for predicting pedestrians crossing behavior (Gu et al, 2016; Keller and Gavrila, 2014; Kooij et al, 2014; Mínguez et al, 2018; Quintero et al, 2014; Roth et al, 2016; Schulz and Stiefelhagen, 2015) and cyclist motion (Pool et al, 2019, 2017; Saleh et al, 2018a; Zernetsch et al, 2016) typically treat each agent individu...…”

Section: Contextual Cuesmentioning

confidence: 99%

“…This category includes several physics-based approaches (Bai et al, 2015; Blaiotta, 2019; Elnagar, 2001; Elnagar and Gupta, 1998; Foka and Trahanias, 2010; Pettré et al, 2009; Schneider and Gavrila, 2013; Zhu, 1991). Pattern-based methods usually model obstacles implicitly, by learning collision-free patterns (Bennewitz et al, 2002; Carvalho et al, 2019; Chen et al, 2016, 2008; Ellis et al, 2009; Ferguson et al, 2015; Han et al, 2019; Hasan et al, 2018; Huynh and Alaghband, 2019; Jacobs et al, 2017; Joseph et al, 2011; Kim et al, 2011; Kruse and Wahl, 1998; Kucner et al, 2017, 2013; Makansi et al, 2019; Makris and Ellis, 2002; Molina et al, 2018; Piciarelli et al, 2005; Saleh et al, 2018a,b; Sun et al, 2018; Sung et al, 2012; Tadokoro et al, 1993; Tay and Laugier, 2008; Thompson et al, 2009; Vasquez et al, 2008; Wang et al, 2015, 2016; Xue et al, 2019, 2017; Yoo et al, 2016). When facing a change in the obstacles’ configuration, such patterns become obstacle-unaware.…”

Section: Contextual Cuesmentioning

confidence: 99%

See 2 more Smart Citations

Human motion trajectory prediction: a survey

Rudenko

Palmieri

Herman³

et al. 2020

The International Journal of Robotics Research

568

306

View full text Add to dashboard Cite

With growing numbers of intelligent autonomous systems in human environments, the ability of such systems to perceive, understand, and anticipate human behavior becomes increasingly important. Specifically, predicting future positions of dynamic agents and planning considering such predictions are key tasks for self-driving vehicles, service robots, and advanced surveillance systems. This article provides a survey of human motion trajectory prediction. We review, analyze, and structure a large selection of work from different communities and propose a taxonomy that categorizes existing methods based on the motion modeling approach and level of contextual information used. We provide an overview of the existing datasets and performance metrics. We discuss limitations of the state of the art and outline directions for further research.

show abstract

Section: Pattern-based Approachesmentioning

confidence: 99%

Section: Contextual Cuesmentioning

confidence: 99%

Section: Contextual Cuesmentioning

confidence: 99%

See 1 more Smart Citation

Human motion trajectory prediction: a survey

Rudenko

Palmieri

Herman³

et al. 2020

The International Journal of Robotics Research

568

306

View full text Add to dashboard Cite

show abstract

“…In addition, long short-term memory (LSTM) was increasingly used as a time recurrent neural network for trajectory prediction [10,11]. Saleh et al [12] compared one-way and two-way LSTM in the trajectory prediction of cyclists. The results showed that the prediction accuracy of both networks is higher than that of traditional MLP.…”

Section: Introductionmentioning

confidence: 99%

Trajectory prediction of cyclist based on dynamic Bayesian network and long short-term memory model at unsignalized intersections

Gao

Cai

et al. 2021

Sci. China Inf. Sci.

View full text Add to dashboard Cite

Cyclist trajectory prediction is of great significance for both active collision avoidance and path planning of intelligent vehicles. This paper presents a trajectory prediction method for the motion intention of cyclists in real traffic scenarios. This method is based on dynamic Bayesian network (DBN) and long short-term memory (LSTM). The motion intention of cyclists is hard to predict owing to potential large uncertainties. The DBN is used to infer the distribution of cyclists' intentions at intersections to improve the prediction time. The LSTM with encoder-decoder is used to predict the cyclists' trajectories to improve the accuracy of prediction. Therefore, the DBN and LSTM are adopted to guarantee prediction accuracy and improve the prediction time. The experiment results are presented to show the effectiveness of the predict strategies.

show abstract

“…Recently, deep learning-based techniques such as convolution neural networks (ConvNets) have been achieving stateof-the-art results in many computer vision tasks such: object identification [1], scene understanding [2], [3], and human action recognition [4]- [6]. However, these techniques require a handful amount of labelled data for training them which is both time-consuming and cumbersome to get for many tasks.…”

Section: Introductionmentioning

confidence: 99%

Domain Adaptation for Vehicle Detection from Bird's Eye View LiDAR Point Cloud Data

Saleh

Abobakr

Attia

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

Self Cite

View full text Add to dashboard Cite

Point cloud data from 3D LiDAR sensors are one of the most crucial sensor modalities for versatile safety-critical applications such as self-driving vehicles. Since the annotations of point cloud data is an expensive and time-consuming process, therefore recently the utilisation of simulated environments and 3D LiDAR sensors for this task started to get some popularity. With simulated sensors and environments, the process for obtaining an annotated synthetic point cloud data became much easier. However, the generated synthetic point cloud data are still missing the artefacts usually exist in point cloud data from real 3D LiDAR sensors. As a result, the performance of the trained models on this data for perception tasks when tested on real point cloud data is degraded due to the domain shift between simulated and real environments. Thus, in this work, we are proposing a domain adaptation framework for bridging this gap between synthetic and real point cloud data. Our proposed framework is based on the deep cycle-consistent generative adversarial networks (CycleGAN) architecture. We have evaluated the performance of our proposed framework on the task of vehicle detection from a bird's eye view (BEV) point cloud images coming from real 3D LiDAR sensors. The framework has shown competitive results with an improvement of more than 7% in average precision score over other baseline approaches when tested on real BEV point cloud images.

show abstract

Cyclist Trajectory Prediction Using Bidirectional Recurrent Neural Networks

Cited by 17 publications

References 15 publications

Human motion trajectory prediction: a survey

Human motion trajectory prediction: a survey

Trajectory prediction of cyclist based on dynamic Bayesian network and long short-term memory model at unsignalized intersections

Domain Adaptation for Vehicle Detection from Bird's Eye View LiDAR Point Cloud Data

Contact Info

Product

Resources

About