Bounding overwatch

About: Bounding overwatch is a research topic. Over the lifetime, 966 publications have been published within this topic receiving 15156 citations.

Action-ViT: Pedestrian Intent Prediction in Traffic Scenes

Abstract: Pedestrian crossing intention prediction is crucial to traffic safety, which is a challenging task in real traffic scenarios. Traditional methods infer the intention of pedestrians to cross by predicting their future movements based on the observed trajectories in history. The performance of those methods is limited due to insufficient features and sources of information. To address those limitations, we propose a ViT-based model which incorporates multi-modal data to predict the pedestrian crossing intention. Specifically, the proposed model takes into consideration the visual information, poses, bounding box coordinates and action annotations, and gradually fuses those features for the final prediction. Besides, different data processing methods are designed based on the corresponding characteristics of different modalities to make full use of each type of data. Extensive ablation studies are conducted to show the performance of temporal modelling and feature fusion.

Bounding filters in the presence of inexactly known parameters

Abstract: Kalman-Bucy (K/B) filtering assumes that the linear dynamic system and corresponding noise statistics are exactly known. When this information is only approximately known, an optimum bounding filter can be designed for specific versions (steady state time-invariant with scalar observations) of the K/B problem. A system is designed in which the actual estimation error covariance is bounded by the covariance calculated by the estimator. Therefore, the estimator obtains a bound on the actual error covariance which is not available and also the apparent divergence is prevented. The bounding filter can be designed to be of lower order than the original system, however, this results in a higher error covariance. Conditions for the design of the optimum (minimum mean square error) bounding filter within a permissible class are derived. Since the specific K/B problem is formulated as an equivalent Wiener filtering problem, therefore, all the results are applicable to the design of bounding filters for Wiener problems. The design of a bounding filter is illustrated by an example.

Dynamic Threshold Optimization - A New Approach?

Abstract: Dynamic Threshold Optimization (DTO) adaptively "compresses" the decision space (DS) in a global search and optimization problem by bounding the objective function from below. This approach is different from "shrinking" DS by reducing bounds on the decision variables. DTO is applied to Schwefel's Problem 2.26 in 2 and 30 dimensions with good results. DTO is universally applicable, and the author believes it may be a novel approach to global search and optimization.

Finite-time state bounding for homogeneous nonlinear positive systems with time-varying delay and bounded disturbance

Abstract: In this paper, the problem of finite-time state bounding for homogeneous nonlinear positive systems with time-varying delay and bounded disturbance is studied for the first time. By exploiting an approach presented in positive systems which is different from the Lyapunov–Krasovskii functional technique, an explicit criterion is formulated for the existence of a special ball such that all the state trajectories of homogeneous nonlinear positive systems with degree 0<α<1 converge within it in finite time. We then extend the main result to the general nonlinear time variant systems. Finally, a numerical example is presented to demonstrate the effectiveness of the theoretical results.