Towards a 3D localization system for an intelligent walker in uneven outdoor environments

The work presented in this paper addresses a practical approach to the problem of 3D pose estimation. The proposed method extends a classical 2D dead reckoning system to a 3D pose estimation system by merging data from odometry and multiple low cost rate gyros and accelerometers. The localization problem is decomposed into two parts, i.e. attitude estimation followed by pose estimation. Based on the innovation sequence, the pitch and roll angles are estimated by an R-adaptive Kalman filter. The adaptive filter is initialized with the maximum measurement noise level resulting from non-gravitational acceleration. Based on the discrepancy between the theoretical and the actual innovation covariance, the measurement covariance R is adjusted by applying a scalar gain for each time step. Heading is calculated based on the gyrodometry algorithm. Finally, the attitude information is fused with data coming from the wheel encoders in order to estimate the 3D position of the robotic walker. Experimental results to investigate the performance of the proposed adaptive Kalman filter and the 3D localization system are presented.