arXiv subjects

Alireza Talebpour

Publications and source records attributed to Alireza Talebpour.

23 records · Page 2Linked to original sources

arXiv2019

Back to the Future: Predicting Traffic Shockwave Formation and Propagation Using a Convolutional Encoder-Decoder Network

This study proposes a deep learning methodology to predict the propagation of traffic shockwaves. The input to the deep neural network is time-space diagram of the study segment, and the output of the network is the predicted (future) propagation of the shockwave on the study segment in the form of time-space diagram. The main feature of the proposed methodology is the ability to extract the features embedded in the time-space diagram to predict the propagation of traffic shockwaves.

cs.LG↗

arXiv2018

A Correlation Based Feature Representation for First-Person Activity Recognition

In this paper, a simple yet efficient activity recognition method for first-person video is introduced. The proposed method is appropriate for representation of high-dimensional features such as those extracted from convolutional neural networks (CNNs). The per-frame (per-segment) extracted features are considered as a set of time series, and inter and intra-time series relations are employed to represent the video descriptors. To find the inter-time relations, the series are grouped and the linear correlation between each pair of groups is calculated. The relations between them can represent the scene dynamics and local motions. The introduced grouping strategy helps to considerably reduce the computational cost. Furthermore, we split the series in temporal direction in order to preserve long term motions and better focus on each local time window. In order to extract the cyclic motion patterns, which can be considered as primary components of various activities, intra-time series correlations are exploited. The representation method results in highly discriminative features which can be linearly classified. The experiments confirm that our method outperforms the state-of-the-art methods on recognizing first-person activities on the two challenging first-person datasets.

cs.CV↗

arXiv2018

Assessing Strong String Stability of Constant Spacing Policy under Speed Limit Fluctuations

The speed limit changes frequently throughout the transportation network, due to either safety (e.g., change in geometry) or congestion management (e.g., speed harmonization systems). Any abrupt reduction in the speed limit can create a shockwave that propagates upstream in traffic. Dealing with such an abrupt reduction in speed limit is particularly important while designing control laws for a platoon of automated vehicles from both stability and efficiency perspectives. This paper focuses on Adaptive Cruise Control (ACC) based platooning under a constant spacing policy, and investigates the possibility of designing a controller that ensures stability, while tracking a given target velocity profile that changes as a function of location. An ideal controller should maintain a constant spacing between successive vehicles, while tracking the desired velocity profile. The analytical investigations of this paper suggest that such a controller does not exist.

eess.SY↗

arXiv2017

Collaborative Platooning of Automated Vehicles Using Variable Time-Gaps

Connected automated vehicles (CAVs) could potentially be coordinated to safely attain the maximum traffic flow on roadways under dynamic traffic patterns, such as those engendered by the merger of two strings of vehicles due a lane drop. Strings of vehicles have to be shaped correctly in terms of the inter-vehicular time-gap and velocity to ensure that such operation is feasible. However, controllers that can achieve such traffic shaping over the multiple dimensions of target time-gap and velocity over a region of space are unknown. The objective of this work is to design such a controller, and to show that we can design candidate time-gap and velocity profiles such that it can stabilize the string of vehicles in attaining the target profiles. Our analysis is based on studying the system in the spacial rather than the time domain, which enables us to study stability as in terms of minimizing errors from the target profile and across vehicles as a function of location. Finally, we conduct numeral simulations in the context of shaping two platoons for merger, which we use to illustrate how to select time-gap and velocity profiles for maximizing flow and maintaining safety.

eess.SY↗

arXiv2017

Platooning in the Presence of a Speed Drop: A Generalized Control Model

The positive impacts of platooning on travel time reliability, congestion, emissions, and energy consumption have been shown for homogeneous roadway segments. However, speed limit changes frequently throughout the transportation network, due to either safety-related considerations (e.g., workzone operations) or congestion management schemes (e.g., speed harmonization systems). These abrupt changes in speed limit can result in shock- wave formation and cause travel time unreliability. Therefore, designing a platooning strategy for tracking a reference velocity profile is critical to enabling end-to-end platooning. Accordingly, this study introduces a generalized control model to track a desired velocity profile, while ensuring safety in the platoon of autonomous vehicles. We define appropriate natural error terms and the target curve in the state space of the control system, which is the set of points where all error terms vanish and corresponds to the case when all vehicles move with the desired velocities and in the minimum safe distance between them. In this way, we change the tracking velocity profile problem into a state- feedback stabilization problem with respect to the target curve. Under certain mild assumptions on the Lipschitz constant of the speed drop profile, we show that the stabilizing feedback can be obtained via introducing a natural dynamics for the maximum of the error terms for each vehicle. Moreover, we show that with this stabilizing feedback collisions will not occur if the initial state of the system of vehicles is sufficiently close to the target curve. We also show that the error terms remain bounded throughout the time and space. Two scenarios were simulated, with and without initial perturbations, and results confirmed the effectiveness of the proposed control model in tracking the speed drop while ensuring safety and string stability.

eess.SY↗