Slip-compensated odometry for tracked vehicle on loose and weak slope

[摘要] Odometry is widely used to localize wheeled and tracked vehicles because of its simplicity and continuity. Odometric calculations integrate the wheel or track’s rotation speed. The accuracy of position thus calculated, is affected by slippage between the ground and the wheel or track. When traveling on a loose slope, the localization accuracy of the odometry decreases remarkably due to slippage. To improve its accuracy in such environments, terramechanics focus on estimating the interaction between a vehicle and the ground. However, because these formulas are complicated and governed by many terrain-specific parameters, they are difficult to use in unknown environments. In this study, we propose slip estimation methods targeted toward use in unknown environments. We consider four types of slippage, based on the slippage direction and maneuver type. Longitudinal and lateral slippage occurring during straight maneuvering are derived by approximating the terramechanics slip model. In contrast, for turning maneuvers, longitudinal slippage is derived from an empirical equation for the relationship between slip ratio and input velocity, and lateral slippage is obtained from a regression function. We also proposed slip-compensated odometry, which applies the slip model to the kinematics of a skid-steering vehicle. To evaluate the proposed slip model and slip-compensated odometry, we conduct several experiments with a skid-steering tracked vehicle on an indoor sandy slope. Experimental results confirmed that position estimation accuracy was improved by introducing slip-compensated odometry compared to conventional odometry.

[学科分类] 人工智能

[关键词] Skid-steering tracked vehicle;Odometry;Slip estimation