Skip to Content

Ship deck simulation for autonomous landing of VTOL RPAS

To design and test a controller for autonomous landing of VTOL UAVs on ships, it is necessary to calculate how the ship moves in the Sea, using the whole six degrees of freedom:

Fig 1: Ship movements. Six degrees of freedom

 The movement of the ship depends on:

  • The Sea State: Ten sea states are defined
Fig 2: Sea States from 0 to 9.
  • The direction of the wave
  • The ship characteristics: Size, shape, weigh, ... of the ship.

For the simulation of the ship on the sea, a modification of the proposal of S.Kantsis; A. Bourmistrova and M.R. Crump is done. We propose to use sinus function for each ship movement with random amplitude between zero and the maximum value for each sea state. To get a derivable movement, between two sinus, an interpolation is done.

With our proposal, the following graphs of the CoG of the ship are obtained:

Fig 3: Position of the CoG of the Ship: Surge (x, blue), Sway (y, red) and Heave (z, green). Fig 4: Euler Angles of the Ship: Roll (blue), Pitch (red) and Yaw (green).

As the Deck is not in the CoG of the ship, it is obtained:

Fig 5: Position of the Center of the Ship Deck Platform: Surge (x, blue), Sway (y, red) and Heave (z, green).

The movement of the ship was compared with data of a real ship, looking both similar.

We simulate the movement of the ship deck on the Sea, using a Servos and Simulation Inc, Generic Motion System (model 710-6-500-220) with a 2.44 x 2.44 m2 gray surface as heliport.

Fig 6: Servos and Simulation Inc, 710-6-500-220 Generic Motion System. Number of axis: 6; Height: 48.6 cm; Floor Platform: 66 x 68.6 cm2; Power: 220 VAC @ 20 A; Payload: 226.8 kg; Max. Roll (x): +-13º; Max. Pitch (y): +-15º; Max. Yaw (z): +-16º; Max. Surge (x): +-10.2 cm; Max. Sway (y): +-10.2 cm; and Max. Heave (z): +-6.4 cm

 Once the ship simulation is calculated, she entire system has to be scaled down. Next, the motor inputs of our platform have to be calculated through the Inverse Kinematics, and being carefully with singular configurations.

Fig. 7: Desired motion of our motion platform. In red, points of singular configuration that are not achievable by our platform; in blue achievable points.
Fig. 8: Motor inputs (Volts) before filtering for the desired movement.

The last step is the filtering of the calculated inputs in order to limit the speeds and accelerations because the inverse kinematics calculation does not take them into account:

Fig. 9: Motor inputs (Volts) after filtering inputs.

 You can see a video of the whole system working in the following video: