Obstacle Detection Based on Occupancy Grid Maps Using Stereovision System

Abstract

We previously reported on an obstacle detection method using a stereovision system. The system generated disparity images that include three-dimensional spatial information. Using these images, obstacles could be detected, but some false positives were generated. In this paper, we attempt to eliminate this problem and propose a method that generates Occupancy Grid Maps based on measurements from a stereovision system which leads to robust obstacle detection. Furthermore, it is confirmed that high distance accuracy can be achieved by using our method.

Fast dynamic object extraction using stereovision based on Occupancy Grid Maps and optical flow

“…In our previous report [7], we proposed a moving object extraction algorithm based on Occupancy Grid Maps (OGM). In this section, we briefly explain about the method.…”

“…Moreover, our previous system suffered from a significant deterioration in distance accuracy as the baseline was shortened. In our previous report [7], for the solution of such problems, we proposed an obstacle detection method using Occupancy Grid Maps (OGM). Since, in this method, the existence of an obstacle was represented as a posterior probability based on all past measurements obtained by the stereovision system from moment to moment, the number of false positives is reduced and the deterioration in distance accuracy that arises from baseline shortening can be ameliorated.…”

Enhancemend of moving object extraction using 6D information

“…For the solution of such problems, we proposed an obstacle detection method [7] using Occupancy Grid Maps (OGM). Since, in this method, the existence of an obstacle was represented as a posterior probability based on all past measurements obtained by the stereovision system from moment to moment, the number of false positives is reduced and the deterioration in distance accuracy that arises from baseline shortening can be ameliorated.…”

“…Since, in this method, the existence of an obstacle was represented as a posterior probability based on all past measurements obtained by the stereovision system from moment to moment, the number of false positives is reduced and the deterioration in distance accuracy that arises from baseline shortening can be ameliorated. Moreover, by comparing time series of generated OGM, moving obstacles were extracted robustly [8]. However, there was some scene where the moving objects are not correctly separated from stationary object because of less distance accuracy.…”

“…In our previous reports [8], we proposed a moving object extraction algorithm based on Occupancy Grid Maps (OGM). In this section, we briefly explain the method.…”