Robotica

Articles

Precise outdoor localization with a GPS–INS integration system

Wonkyo Seoa1, Seoyoung Hwanga1, Jaehyun Parka1 and Jang-Myung Leea1 c1

a1 School of Electrical Engineering, Pusan National University, Busan, South Korea

SUMMARY

This paper proposes a precise outdoor localization algorithm with the integration of Global Positioning System (GPS) and Inertial Navigation System (INS). To achieve precise outdoor localization, two schemes are recently proposed, which consist of de-noising the INS signals and fusing the GPS and INS data. To reduce the noise from the internal INS sensors, the discrete wavelet transform and variable threshold method are utilized, and to fuse the GPS and INS data while filtering out the noise caused by the acceleration, deceleration, and unexpected slips, the Unscented Particle Filter (UPF) is adopted. Conventional de-noising methods mainly employ a combination of low-pass and high-pass filters, which results in signal distortion. This newly proposed system also utilizes the vibration information of the actuator according to the fluctuations of the velocity to minimize the signal distortion. The UPF resolves the nonlinearities of the actuator and non-normal distributions of the noise more effectively than the conventional particle filter (PF) or Extended Kalman Filter–PF. The superiority of the proposed algorithm was verified through experiments, and the results are reported.

(Accepted July 02 2012)

(Online publication July 26 2012)

KEYWORDS:

  • Robot localization;
  • Mobile robots;
  • Control of robotic systems;
  • Motion planning;
  • Navigation

Correspondence

c1 E-mail: jmlee@pusan.ac.kr