1

MAXX WILSON | MELISSA CRUZ

DECEMBER 2021

Unmanned Underwater Vehicle Controller

ME397 Digital Control

Objectives

• Design Attitude Controller using emulation for existing vehicle model
• Examine effect of discretization and sampling frequency on controller

2

The Pioneer REMUS-600 Autonomous Underwater Vehicle. (Source: White 2020)

Vehicle Modeling

3

Adapted from Engelhardtsen (2007) and Fossen (2011)

Parameters from da Silva (2007)

Problem Insights

• High damping allows for use of derivative in controller
• Emulation is suitable technique given slow dynamics compared to hardware speed

4

Continuous Controller Design - Roll

• For surveying and linearization assumptions, roll regulated to zero
• Small variations in COG can cause steady state error
• Eliminate SS error using integral controller (Ki = 1.0)
• Pure Integral can lead to instability if model is inaccurate

5

Continuous Controller Design - Roll

6

Continuous Controller Design - Pitch

PID controller (Prestero, 2001)

• 0.75s settling time
• Overshoot limited to 5%
• Zero Steady-State Error

Gains:

• Kp = 9.34
• Ki = 8.57
• Kd = 1.98

Response:

• Settling Time of 0.7s
• Overshoot of 4.86%

7

Continuous Controller Design - Yaw

PD Controller (Prestero, 2001)

• 0.75s settling time
• Overshoot limited to 5%

Gains

• Kp = 9.22
• Kd = 1.79

Response

• Settling Time of 0.754s
• Overshoot of 4.98%

8

Continuous Controller Design - Yaw

9

Time (s)

Time (s)

Closed Loop Yaw Tracking for 0.1 Hz input

Commanded Fin Angle (deg)

Digital Controller - Sampling Time

Using emulation, we can obtain a discretized controller and vary sampling time to evaluate controller accuracy

​

The following slides show the discrete controller response obtained using ZOH

​

​

10

11

Ts = 0.1 s | Sampling at 10Hz

12

Ts = 0.02s | Sampling at 50Hz

13

Ts = 0.004s | Sampling at 250Hz

Digital Controller - Discretization Method

• Similarly, we can illustrate the effect of variations of emulation methods on discretization performance
• Vehicle dynamics are sufficiently lower than nyquist frequency when sampling at 50Hz
• ZOH method performs well over required frequency range

14

Bode Plot of Closed Loop Yaw Response at Ts = 0.02s

Digital Controller - Low Pass Filter

• Actuator saturation limited effective bandwidth to 1.5 rad/s in simulation
• Filter coefficient can be set to 10 rad/s to reduce sensor noise effects

15

Conclusions

• Emulation is suitable for our application since the hardware runs much faster than vehicle dynamics, especially using a sampling frequency of 50Hz
• Steady state error requirement and high stiffness motivate a pure Integral controller in Roll
• Pitch and Yaw are motivated by time criterion of 0.75s settling time, overshoot limited to 5%, and zero Steady-State Error
• Yaw control is realized using PD control
• Pitch control meets criteria with full PID control

​

16

Future Work

• Explore control allocation methods to handle fin saturation
• Incorporate position tracking algorithm to follow paths in XY-plane
• Explore observer models to track unmeasured variables (important for waypoint navigation)

17

References

Breivik, Morten & Fossen, Thor. (2009). Guidance Laws for Autonomous Underwater Vehicles. 10.5772/6696.

Fossen, T. (2002), Marine Control Systems, Marine Cybernetics AS.

Fossen, T.I. (2011). Models for Ships, Offshore Structures and Underwater Vehicles. In Handbook of Marine Craft Hydrodynamics and Motion Control, T.I. Fossen (Ed.). https://doi.org/10.1002/9781119994138.ch7

Manhães, M.M., Scherer, S.A., Voss, M., Douat, L.R., & Rauschenbach, T. (2016). UUV Simulator: A Gazebo-based package for underwater intervention and multi-robot simulation. OCEANS 2016 MTS/IEEE Monterey, 1-8.

da Silva, Jorge Estrela, et al. (2007). Modeling and Simulation of the LAUV autonomous underwater vehicle. 13th IEEE IFAC International Conference on Methods and Models in Automation and Robotics. Szczecin, Poland.

Engelhardtsen, Ø. (2007). 3D AUV Collision Avoidance [Thesis]. Norwegian University of Science and Technology.

Prestero, T. (2001). Verification of a Six-Degree of Freedom Simulation Model for the Remus Autonomous Underwater Vehicle [Thesis]. Massachusetts Institute of Technology.

White, S. (2020). “The Pioneer REMUS-600 Autonomous Underwater Vehicle.” Robotics AUVs, Ocean Observatories Initiative, https://oceanobservatories.org/marine-technologies/robotic-auvs/.

Yang, C. (2007). Modular Modelling and Control for Autonomous Vehicle (AUV) [Thesis]. Department of Mechanical Engineering National University of Singapore.

18

Questions?

19