Trajectory prediction of cyclists using a physical model and an artificial neural network

Zernetsch, Stefan; Kohnen, Sascha; Goldhammer, Michael; Doll, Konrad; Sick, Bernhard

doi:10.1109/ivs.2016.7535484

Cited by 70 publications

(39 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar studies are also being performed for cyclists. Zernetsch et al (2016) collected data at a single intersection for path prediction of a starting cyclists, and Hubert et al (2017) used the same data to find indicators of cyclist starting behavior. Some studies have used naturalistic data to detect and classify critical vehicle-cyclist interactions at intersections (Sayed et al 2013;Vanparijs et al 2015;Cara and de Gelder 2015), while others use simulations to study bicycle motion at intersections Zhang et al 2017).…”

Section: Context Cues For Vru Behaviorsmentioning

confidence: 99%

Context-Based Path Prediction for Targets with Switching Dynamics

et al. 2018

View full text Add to dashboard Cite

Anticipating future situations from streaming sensor data is a key perception challenge for mobile robotics and automated vehicles. We address the problem of predicting the path of objects with multiple dynamic modes. The dynamics of such targets can be described by a Switching Linear Dynamical System (SLDS). However, predictions from this probabilistic model cannot anticipate when a change in dynamic mode will occur. We propose to extract various types of cues with computer vision to provide context on the target's behavior, and incorporate these in a Dynamic Bayesian Network (DBN). The DBN extends the SLDS by conditioning the mode transition probabilities on additional context states. We describe efficient online inference in this DBN for probabilistic path prediction, accounting for uncertainty in both measurements and target behavior. Our approach is illustrated on two scenarios in the Intelligent Vehicles domain concerning pedestrians and cyclists, so-called Vulnerable Road Users (VRUs). Here, context cues include the static environment of the VRU, its dynamic environment, and its observed actions. Experiments using stereo vision data from a moving vehicle demonstrate that the proposed approach results in more accurate path prediction than SLDS at the relevant short time horizon (1 s). It slightly outperforms a computationally more demanding state-of-the-art method.

show abstract

Section: Context Cues For Vru Behaviorsmentioning

confidence: 99%

Context-Based Path Prediction for Targets with Switching Dynamics

et al. 2018

View full text Add to dashboard Cite

show abstract

“…The keys to the UAV's early warn ing for the no-fly zone are the prediction of UAV flight trajectory [7][8] and the difficu lty of how to determine whether the UAV will enter the no -fly zone. Trajectory prediction was presented for the movement track of moving target by the linear neural network [9][10] . The paper compared the algorith m based on α/β/γ filtering, Kalman filtering(KF), and interacting multiple [11] .An improved KF was proposed to estimate the 4D trajectory [12] , which increased the accuracy of trajectory prediction through real-time estimat ion of system noise, but large number of training samples, and a slow learning speed existed.…”

Section: Related Workmentioning

confidence: 99%

NUCLEI: A New Unmanned Aerial Vehicle Early Warning Method for the No-fly Zone

Wang¹,

Yu²,

Ma³

et al. 2018

Proceedings of the 2018 International Conference on Advanced Control, Automation and Artificial Intelligence (ACAAI 2018)

View full text Add to dashboard Cite

Abstract-In order to monitor the unmanned aerial vehicle(UAV) around the no-fly zone, we proposed a new unmanned aerial vehicle Early warning method for the no-fly zone (NUCLEI) which gives an alarm to the UAV around the nofly zone in real time. We utilize the least square method to match with the flight trajectory of the UAV in the current period, and then we use the minimum external rectangle (MERT) method to obtain the MERT of the electronic fence. Finally, we calculate whether the tangent line at the current point of the flight path fits curve will intersect with the electronic fence to make the prediction for if the UAV will enter the no-fly zone.

show abstract

“…There has been work on how cyclist motion evolves at a crossing [7], [8], but this was from static viewpoints outside of the vehicle. Furthermore, [19] proposed a novel datasets for learning social dynamics in crowded scenes, which in addition to 11216 pedestrians includes 6364 cyclist tracks [19].…”

Section: Related Workmentioning

confidence: 99%

“…While research on pedestrian path prediction yielded various publications and datasets, relatively few work currently focuses on cyclist [7], [8]. However, recently a large benchmark dataset on cyclist detection from a moving vehicle was made publicly available [9].…”

Section: Introductionmentioning

confidence: 99%

Using road topology to improve cyclist path prediction

Pool

Kooij

Gavrila

2017

2017 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Abstract-We learn motion models for cyclist path prediction on real-world tracks obtained from a moving vehicle, and propose to exploit the local road topology to obtain better predictive distributions. The tracks are extracted from the Tsinghua-Daimler Cyclist Benchmark for cyclist detection, and corrected for vehicle egomotion. Tracks are then spatially aligned to local curves and crossings in the road. We study a standard approach for path prediction in the literature based on Kalman Filters, as well as a mixture of specialized filters related to specific road orientations at junctions. Our experiments demonstrate an improved prediction accuracy (up to 20% on sharp turns) of mixing specialized motion models for canonical directions, and prior knowledge on the road topology. The new track data complements the existing video, disparity and annotation data of the original benchmark, and will be made publicly available.

show abstract

Trajectory prediction of cyclists using a physical model and an artificial neural network

Cited by 70 publications

References 11 publications

Context-Based Path Prediction for Targets with Switching Dynamics

Context-Based Path Prediction for Targets with Switching Dynamics

NUCLEI: A New Unmanned Aerial Vehicle Early Warning Method for the No-fly Zone

Using road topology to improve cyclist path prediction

Contact Info

Product

Resources

About