The standard vector phase-locked loop (VPLL) with fixed pre-defined process noise covariance matrix usually loses lock when the receiver experiences high dynamic conditions. Two adaptive Kalman filters (KF), widely used for position calculation in Global Navigation Satellite System (GNSS) receivers, and their characteristics are analyzed and considered as potential solutions to this issue. An adaptive joint VPLL is then proposed and implemented to improve phase tracking performance in high dynamic conditions. In the proposed adaptive joint VPLL, the outputs of the individual filters in the joint VPLL are used to adaptively adjust the process noise covariance matrix of the common filter based on an Extended KF (EKF). A hardware simulator test and a field test have been conducted to assess the performance of the proposed adaptive joint VPLL, which is compared with the conventional VPLL. The test results show that the adaptive joint VPLL can improve the robustness of carrier phase tracking in high dynamic conditions.