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

Goldhammer, Michael; Köhler, Sebastian; Zernetsch, Stefan; Doll, Konrad; Sick, Bernhard; Dietmayer, Klaus

doi:10.1109/tits.2019.2923319

Cited by 49 publications

(22 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After smoothing the original signal, the related feature quantity is extracted from the time and the frequency domains [25]. In order to make full use of the information in the original signal, 14 eigenvalues are extracted from 9 original signals, such as triaxial acceleration, triaxial angular velocity, and attitude angle of three axes, respectively, which are the mean value of acceleration, the mean value of synthetic acceleration, the variance of synthetic acceleration, the peak valley value of synthetic acceleration, the average synthetic angular velocity, the variance of synthetic angular velocity, the peak valley value of synthetic angular velocity, the covariance of synthetic acceleration and synthetic angular velocity, the change of attitude angle in X, Y, and Z directions in the time domain, and the energy value of acceleration signal in the frequency domain.…”

Section: Feature Extractionmentioning

confidence: 99%

Remote Evaluation System of Tennis Batting Action Standard Based on Acceleration Sensor

2021

Security and Communication Networks

View full text Add to dashboard Cite

This paper studies the remote evaluation system of tennis batting action standard based on acceleration sensor, which aims to help improve the standard degree and technical level of tennis batting action. The system includes a data acquisition module to collect original signal data of tennis batting action by the acceleration sensor signal acquisition device in the bracelet and upload to the personal computer (PC) for storage, data preprocessing module to smooth original signal data and extract the key time and frequency domain features as the evaluation basis, and remote evaluation module to assess tennis batting action standard. We applied our system to five tennis trainees from the experimental university, and the results show that the batting action standard level of student c and student e is lower. Student c is weak mainly in the best position of the hitting point and the timing of the lead shot, while student e mainly shows poor performance in the timing of movement and the stability of the overall center of gravity. Compared with the proposed system or device, our system has a short real-time delay under the concurrent use of different types of users indicating stable and high real-time evaluation performance. More importantly, our system strictly protects the user’s privacy when uploading the user’s data remotely. In short, the evaluation results obtained by our system can be used as a scientific basis to improve the tennis batting action standard.

show abstract

Section: Feature Extractionmentioning

confidence: 99%

Remote Evaluation System of Tennis Batting Action Standard Based on Acceleration Sensor

2021

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…The motion state of pedestrian has lesser impact on the restrictions of traffic signs and rules. So, the shortterm trajectory prediction task of pedestrians is more challenging [2]. Interaction among pedestrians is an important aspect that affects trajectory.…”

Section: Related Workmentioning

confidence: 99%

“…The auxiliary factor is calculated based on the differences between the first node and other nodes in the trajectory sequence, as shown in (2). A represents the sequence of different auxiliary factors.…”

Section: Auxiliary Factors and Interactivity Among Pedestriansmentioning

confidence: 99%

DeepPTP: A Deep Pedestrian Trajectory Prediction Model for Traffic Intersection

Lv¹,

Li²,

Dong³

et al. 2021

KSII TIIS

View full text Add to dashboard Cite

Compared with vehicle trajectories, pedestrian trajectories have stronger degrees of freedom and complexity, which poses a higher challenge to trajectory prediction tasks. This paper designs a mode to divide the trajectory of pedestrians at a traffic intersection, which converts the trajectory regression problem into a trajectory classification problem. This paper builds a deep model for pedestrian trajectory prediction at intersections for the task of pedestrian shortterm trajectory prediction. The model calculates the spatial correlation and temporal dependence of the trajectory. More importantly, it captures the interactive features among pedestrians through the Attention mechanism. In order to improve the training speed, the model is composed of pure convolutional networks. This design overcomes the single-step calculation mode of the traditional recurrent neural network. The experiment uses Vulnerable Road Users trajectory dataset for related modeling and evaluation work. Compared with the existing models of pedestrian trajectory prediction, the model proposed in this paper has advantages in terms of evaluation indicators, training speed and the number of model parameters.

show abstract

“…With the rise of IoT devices, the idea of edge computing is also gaining prominence and is broadly recognized. Machine learning solutions provide improved guard safety in hazardous traffic circumstances in the context of traffic applications [5][6][7].…”

Section: Introductionmentioning

confidence: 99%

Privacy-Aware Data Forensics of VRUs Using Machine Learning and Big Data Analytics

Babar

Tariq

Almasoud

et al. 2021

Security and Communication Networks

View full text Add to dashboard Cite

The present spreading out of big data found the realization of AI and machine learning. With the rise of big data and machine learning, the idea of improving accuracy and enhancing the efficacy of AI applications is also gaining prominence. Machine learning solutions provide improved guard safety in hazardous traffic circumstances in the context of traffic applications. The existing architectures have various challenges, where data privacy is the foremost challenge for vulnerable road users (VRUs). The key reason for failure in traffic control for pedestrians is flawed in the privacy handling of the users. The user data are at risk and are prone to several privacy and security gaps. If an invader succeeds to infiltrate the setup, exposed data can be malevolently influenced, contrived, and misrepresented for illegitimate drives. In this study, an architecture is proposed based on machine learning to analyze and process big data efficiently in a secure environment. The proposed model considers the privacy of users during big data processing. The proposed architecture is a layered framework with a parallel and distributed module using machine learning on big data to achieve secure big data analytics. The proposed architecture designs a distinct unit for privacy management using a machine learning classifier. A stream processing unit is also integrated with the architecture to process the information. The proposed system is apprehended using real-time datasets from various sources and experimentally tested with reliable datasets that disclose the effectiveness of the proposed architecture. The data ingestion results are also highlighted along with training and validation results.

show abstract

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

Cited by 49 publications

References 36 publications

Remote Evaluation System of Tennis Batting Action Standard Based on Acceleration Sensor

Remote Evaluation System of Tennis Batting Action Standard Based on Acceleration Sensor

DeepPTP: A Deep Pedestrian Trajectory Prediction Model for Traffic Intersection

Privacy-Aware Data Forensics of VRUs Using Machine Learning and Big Data Analytics

Contact Info

Product

Resources

About