Innovative adaptive autopilot design for uninhabited surface vehicles

In real life situations, very often a sudden change in dynamics results in missions being aborted which results in it having to be rescued before damage to other marine craft in the vicinity is caused. This problem has been suitably dealt with by an innovative adaptive autopilot design that is described herein. A model predictive control (MPC) adopts an online adaptive nature by utilizing weighted least squares (WLS). Even with random initialization, significant improvements were achieved by WLS by maintaining the intermittent continuous values of system parameters and periodically reinitializing the covariance matrix. Also, a time frame of 25 seconds appears to be the optimum to reinitialize the parameters. This novel approach enables the autopilot to cope well with significant changes in the system dynamics and empowers USVs to accomplish their desired missions.