Appointed-Time Prescribed Performance Control for Shipborne SideArm With DRL-Based Kinematic Compensation

During the recovery of fixed-wing autonomous aerial vehicles (AAVs) by a SideArm, the motion of the desired docking point (DDP) on the arresting cable located at the SideArm’s tip is heavily influenced by the ship’s movements. The movements increase the complexity and risk of the docking procedure. As a means of stabilizing the spatial position of the DDP, this paper presents a novel control framework that integrates kinematic solutions and dynamic control strategies. The proposed framework first compensates for the ship’s six-degrees-of-freedom motion, then generates the basic joint signals of the SideArm using a geometry-based inverse kinematic (IK) method. The twin delayed deep deterministic policy gradient (TD3) algorithm is applied to compensate for these signals, forming the IK-TD3 approach that enhances the accuracy of the DDP’s spatial position. An appointed-time prescribed performance control method is designed to ensure that the system’s performance remains within predefined constraints and converges within a preassigned time. A fixed-time disturbance observer is introduced to estimate the lumped disturbances and uncertainty dynamics of the SideArm. A comprehensive set of simulations and experiment considering different sea states demonstrates the reliability of the proposed approach. Note to Practitioners—This paper is motivated by the problem of recovering AAVs in unpredictable sea conditions using a SideArm system located on a ship’s deck. From a deployment perspective, a fixed-size SideArm mitigates the need to resize equipment based on wave height, but simultaneously increases the difficulty of finding accurate and reliable IK solutions, especially in the case of large ${}^{\mathrm { }}$ amplitude deck motion. Traditional geometric-based IK methods are computationally efficient, but struggle to maintain sufficient accuracy under these conditions. To overcome this limitation, we propose an IK framework that integrates the TD3 algorithm, thus enhancing the SideArm’s ability to compensate for varying ship motions. We improve the controller design using prescribed performance control, thereby ensuring both transient and steady-state performance without requiring an explicit residual set calculation. Numerical simulations and experiments under various sea states demonstrate that the proposed approach achieves robust, high-accuracy joint angle commands and improved system stability, making it a practical and reliable solution for AAV recovery operations in maritime environments.