Efficient calibration of four wheel industrial AGVs

Abstract In this paper, we propose a novel method for extrinsic and intrinsic automatic calibration of four wheel industrial Automated Guided Vehicles (AGVs) compliant with Ackermann and Dual Drive kinematics. For each kinematic model the algorithm estimates the trajectories measured by an on-board sensor and the expected ones given the state of the wheels. The estimation exploits the model equations derived in this work which constrain calibration parameters and measurements from wheel encoders and sensor odometry. The parameter values are computed through closed-form solutions of least-square estimation. The method has been implemented on Programmable Logic Controllers and tested on industrial AGVs. The developed procedure computes the parameters in about 10 − 15 minutes, a significant improvement compared with one hour or more required by manual AGV calibration. Experiments with AGVs of various sizes in a warehouse have assessed the accuracy and stability of the proposed approach. The position accuracy achieved by AGVs calibrated with the proposed method is higher than the one achieved by manual calibration.