Monocular visual odometry from frame to frame intensity differences for planetary exploration mobile robots

Abstract

Traditionally stereo visual odometry algorithms estimate the robot's motion by maximizing the conditional probability of the 3D correspondences between two sets of 3D feature point positions, which were previously obtained from two consecutive stereo image pairs captured by a stereo video camera. As an alternative, in this paper a monocular visual odometry algorithm is proposed, which estimates the robot's motion by maximizing the conditional probability of the frame to frame intensity differences at observation points between two consecutive images captured by a monocular video camera. Experimental results with synthetic and real image sequences revealed highly accurate and reliable estimates, respectively. Additionally, it seems to be an excellent candidate for mobile robot missions where space, weight and power supply are really very limited.