Autonomous landing on the deck of an unmanned surface vehicle (USV) is still a major challenge for unmanned aerial vehicles (UAVs). In this paper, a fiducial marker is located on the platform so as to facilitate the task since it is possible to retrieve its six-degrees of freedom relative-pose in an easy way. To compensate interruption in the marker’s observations, an extended Kalman filter (EKF) estimates the current USV’s position with reference to the last known position. Validation experiments have been performed in a simulated environment under various marine conditions. The results confirmed that the EKF provides estimates accurate enough to direct the UAV in proximity of the autonomous vessel such that the marker becomes visible again. Using only the odometry and the inertial measurements for the estimation, this method is found to be applicable even under adverse weather conditions in the absence of the global positioning system.

This paper proposes a new set-membership technique to implement polytopic set computation for nonlinear discrete-time systems indirectly. The proposed set-membership technique is applied to solve the guaranteed state estimation problem for nonlinear discrete-time systems with a bounded description of noises and parameters. A common practice for this problem in the literature is to search an optimal zonotope to bound the intersection of the evolved uncertain state trajectory and the region of the state space consistent with the observed output at each observation update. The new approach keeps the polytopic set resulting from the intersection intact and computes the evolution of this intact polytopic set for the next time step through representing the polytopic set exactly by the intersection of zonotopes. Such an approach avoids the overapproximation bounding process at each observation update, and thus, a more accurate state estimation can be obtained. An illustrative example is provided to demonstrate the effectiveness of the proposed guaranteed state estimation via indirectly implemented polytopic set computation.

Summary: Autonomous landing on the deck of a boat or an unmanned surface vehicle (USV) is the minimum requirement for increasing the autonomy of water monitoring missions. This paper introduces an end-to-end control technique based on deep reinforcement learning for landing an unmanned aerial vehicle on a visual marker located on the deck of a USV. The solution proposed consists of a hierarchy of Deep Q-Networks (DQNs) used as high-level navigation policies that address the two phases of the flight: the marker detection and the descending manoeuvre. Few technical improvements have been proposed to stabilize the learning process, such as the combination of vanilla and double DQNs, and a partitioned buffer replay. Simulated studies proved the robustness of the proposed algorithm against different perturbations acting on the marine vessel. The performances obtained are comparable with a state-of-the-art method based on template matching.

Incorporating virtual metrology (VM) into run-to-run (R2R) control enables the benefits of R2R control to be maintained while avoiding the negative cost and cycle time impacts of actual metrology. Due to the potential for prediction errors from VM models, the prediction as well as the corresponding confidence information on the predictions should be properly considered in VM-enabled R2R control schemes in order to guarantee control performance. This paper proposes the use of Gaussian process regression (GPR) models in VM-enabled R2R control due to their ability to provide this information in an integrated fashion. The mean value of the GPR prediction is treated as the VM value and the variance of the GPR prediction is used as a measure of confidence to adjust the coefficient of an exponentially weighted-moving-average R2R controller. The effectiveness of the proposed GPR VM-enabled R2R control approach is demonstrated using a chemical mechanical polishing process case study. Results show that better control performance is achieved with the proposed methodology than with implementations that do not take prediction reliability into account.