Tampere University of Technology

TUTCRIS Research Portal

Situation Awareness Framework for Multi-machine Environments: An Application to Operator Assistive Collision Warning System

Research output: Contribution to journalArticleScientificpeer-review

Details

Original languageEnglish
Pages (from-to)336-345
Number of pages10
JournalIEEE Transactions on Intelligent Transportation Systems
Volume17
Issue number2
Early online date2015
DOIs
Publication statusPublished - 2016
Publication typeA1 Journal article-refereed

Abstract

Mobile multi-machine environments consist of varying types of objects, either static or dynamic with the state known exactly or with some uncertainty. Sensors observe the environment from different positions and views, such as horizontally from the top of a mobile machine or vertically downward from an external observing machine. In this paper, we propose a simple grid-based framework for representing the information of varying types of objects in a 2-D environment and Bayesian methods for updating this information through observations and prediction models. This information about the current and near-future state of the environment is called situation awareness (SA). SA information can be utilized as the basis of the operator assistance system for enhancing the safety and efficiency of manual and semiautonomous multi-machine work environments. SA information can also be utilized in task planning of autonomous machines, via creating a dynamic costmap for path planning or entropy map for planning the optimal use of sensory systems. This paper aims at real-time mobile multi-machine environments; hence, the SA framework is kept simple for computational feasibility, but is also general enough to be applicable in other environments as well. Discussion on discretization errors and computational complexity are also covered.

Keywords

  • Observers, Planning, Predictive models, Probabilistic logic, Safety, Uncertainty, Vehicle dynamics, Situation awareness, collision avoidance, driver assistance, multi-machine systems, state prediction

Publication forum classification

Field of science, Statistics Finland