Kalman filter with time-variable gain for a multisensor fusion system

Abstract

In this paper, a development of the sensor fusion algorithm for a visual control of mobile robot is presented. The visual sensor gives many kinds of valuable information and is especially important for the autonomously controlled mobile robots. The output data from the visual sensor include a time-lag due to the image processing computation. Moreover, in most cases, the sampling rate of the uisual sensor is considerably low so that it should be used with other sensors to control fast motion. The development of an algorithm to integmte the dota from multi-sensor system to obtain the optimal estimate of the state variables is a very important problem. The main purpose of this paper is to develop a method which constitutes a sensor fusion system to give the optimal state estimates. The proposed SenSOT fusion system combines the visual sensor and inertial sensor using a modified Kalman filter. A kind of multi-rate Ifalman filter ,which treats the slow sampling rate data from visual sensor is applied for the construction of the sensor fusion system. A n important feature of the multi-mte version of Kalman filter is the use of time-varying filter gain matrix.An expenmental mobile robot is developed which realizes a sensor fusion system combining a vision sensor system and an inertial sensor system. Experimental results show that the proposed sensor fusion algorithm gives the good estimates of the state variables combining the data from CCD camem images and the inertial sensors.

Fusion of IMU and Vision for Absolute Scale Estimation in Monocular SLAM

“…Because no translation was recovered, no scale factor had to be estimated. Also the authors of [12] and [13] only included orientation. In [14] they used a single camera to track lane boundaries on a street for autonomous driving.…”

“…Another is to weight the uncertainty of the measurement according to its temporal occurrence. We used the simplification in [12]. We claim no certainty at all if no measurement is available (i.e.…”

Infrastructureless approach for ubiquitous user location tracking in construction environments

“…The second one consists of weighting the uncertainty of each measurement based on its chronological occurrence. In our system, the simplification described in [58] was implemented. No certainty is claimed till a measurement becomes available.…”

Multi-Sensor, Adjustable-Period Integrated Navigation Method Based on Multi-Stage Signal Trigger for Underwater Vehicles

“…The variations in periods of different navigation sensors can lead to a decline in the precision of an INS, and require improvement. Real-time adjustment of the update criteria of the measurement and the gain matrix of a Kalman filter are adopted in the works of Zhao (2009), Niwa et al (1999) and Wang et al (2010). According to the uncertainty and random and finite measurement errors of Geographic Information System (GIS) information, an in coordinate interval filtering algorithm was proposed to improve the real-time function of a Kalman filter's output signals in Luo et al (2013).…”