Enhanced positioning systems using optical mouse sensors

Abstract

This paper explores the use of optical mouse sensors for building positioning systems. A low-cost, customized optical sensor positioning system has been developed, with desirable high performance. Unlike the traditional dead-reckoning method using rotary encoders, optical mouse sensors are unacted on the errors induced by wheel slippage. However, some research has indicated that height variations can bring errors in 2D displacement measurements when using optical mouse sensors. What’s more, the so-called error cumulative phenomenon is an inevitable problem to every motion positioning system. To overcome these weaknesses, our main contributions are in two aspects. One is that a new adaptation mechanism is presented to enhance the positioning robustness to surface height variations. This performance enhancement attributes to the height-variation awareness of our presented system, in addition to the required positioning capability. The other is that a novel calibration paradigm with the assisted landmarks is proposed to eliminate the error cumulative phenomenon. This induced performance enhancement benefits from the high-speed optic flow sensing capability of optical mouse sensors. Experimental results are reported to validate the enhancement capability of the presented positioning system.