Pedestrian's Trajectory Forecast in Public Traffic with Artificial Neural Networks

Goldhammer, Michael; Doll, Konrad; Brunsmann, U.; Gensler, André; Sick, Bernhard

doi:10.1109/icpr.2014.704

Cited by 26 publications

(27 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…IV). As anchor point for tracking in 3D world space serves the VRUs' centers of gravity (COG) determined from the laser scanner generated point cloud, or the center of the head detected from video data (see [18]), respectively. As changes of the motion state, especially the begin of the pedestrian starting motion, are initialized by a slight upper body bending leading to a shift of the COG into the direction of movement [24], the head movement can serve as an early indicator for the intention.…”

Section: A Time Series Extractionmentioning

confidence: 99%

“…As shown in [18], the time series extracted within a certain time window can directly serve as input pattern of a machine learning predictor. However, we proceed to a further level of abstraction by using the coefficients of an orthogonal expansion of an approximating polynomial as presented in [20].…”

Section: B Representation With Approximating Polynomialsmentioning

confidence: 99%

“…A trajectory prediction method based on two physical models of the pedestrian starting motion is published in [16], corresponding analyses for stopping motions in [17]. A machine learning model using the time series of the pedestrian's ego velocity as input pattern for Multi Layer Perceptrons (MLP) is presented in [18]. This concept is extended with a polynomial representation by least-squares approximation of the input and output time series, leading to a reduction of the feature space dimensionality and an increase of generalisability [19].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Intentions of Vulnerable Road Users—Detection and Forecasting by Means of Machine Learning

Goldhammer

Köhler

Zernetsch

et al. 2020

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

Avoiding collisions with vulnerable road users (VRUs) using sensor-based early recognition of critical situations is one of the manifold opportunities provided by the current development in the field of intelligent vehicles. As especially pedestrians and cyclists are very agile and have a variety of movement options, modeling their behavior in traffic scenes is a challenging task. In this article we propose movement models based on machine learning methods, in particular artificial neural networks, in order to classify the current motion state and to predict the future trajectory of VRUs. Both model types are also combined to enable the application of specifically trained motion predictors based on a continuously updated pseudo probabilistic state classification. Furthermore, the architecture is used to evaluate motion-specific physical models for starting and stopping and video-based pedestrian motion classification. A comprehensive dataset consisting of 1068 pedestrian and 494 cyclist scenes acquired at an urban intersection is used for optimization, training, and evaluation of the different models. The results show substantial higher classification rates and the ability to earlier recognize motion state changes with the machine learning approaches compared to interacting multiple model (IMM) Kalman Filtering. The trajectory prediction quality is also improved for all kinds of test scenes, especially when starting and stopping motions are included. Here, 37% and 41% lower position errors were achieved on average, respectively.

show abstract

Section: A Time Series Extractionmentioning

confidence: 99%

Section: B Representation With Approximating Polynomialsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Intentions of Vulnerable Road Users—Detection and Forecasting by Means of Machine Learning

Goldhammer

Köhler

Zernetsch

et al. 2020

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Most studies featuring the forecasting of pedestrian movements design and train models that forecast pedestrian trajectories (e.g., [26,27,28,29]). Generally, these algorithms derive the current movement behavior of a pedestrian by means of computer vision algorithms.…”

Section: Contemporary Model- and Data Driven (Crowd) Forecasting Mmentioning

confidence: 99%

Forecasting Pedestrian Movements Using Recurrent Neural Networks: An Application of Crowd Monitoring Data

Duives

Wang

Kim

2019

Sensors

View full text Add to dashboard Cite

Currently, effective crowd management based on the information provided by crowd monitoring systems is difficult as this information comes in at the moment adverse crowd movements are already occurring. Up to this moment, very little forecasting techniques have been developed that predict crowd flows a longer time period ahead. Moreover, most contemporary state estimation methods apply demanding pre-processing steps, such as map-matching. The objective of this paper is to design, train and benchmark a data-driven procedure to forecast crowd movements, which can in real-time predict crowd movement. This procedure entails two steps. The first step comprises of a cell sequence derivation method that allows the representation of spatially continuous GPS traces in terms of discrete cell sequences. The second step entails the training of a Recursive Neural Network (RNN) with a Gated Recurrent Unit (GRU) and six benchmark models to forecast the next location of pedestrians. The RNN-GRU is found to outperform the other tested models. Some additional tests of the ability of the RNN-GRU to forecast illustrate that the RNN-GRU preserves its predictive power when a limited amount of data is used from the first few hours of a multi-day event and temporal information is incorporated in the cell sequences.

show abstract

“…Recent advances in modeling human behavior using machine learning techniques have allowed us to reach relatively accurate results for the short-term horizon [12,13]. Unfortunately, many of the above-mentioned applications require long-term prediction.…”

Section: Introductionmentioning

confidence: 99%

Long-term path prediction in urban scenarios using circular distributions

Coscia

Castaldo

Palmieri

et al. 2018

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

Pedestrian's Trajectory Forecast in Public Traffic with Artificial Neural Networks

Cited by 26 publications

References 16 publications

Intentions of Vulnerable Road Users—Detection and Forecasting by Means of Machine Learning

Intentions of Vulnerable Road Users—Detection and Forecasting by Means of Machine Learning

Forecasting Pedestrian Movements Using Recurrent Neural Networks: An Application of Crowd Monitoring Data

Long-term path prediction in urban scenarios using circular distributions

Contact Info

Product

Resources

About