Experiments of Dynamic Control of a 6-DOF AUV

In this chapter some experimental results on dynamic control of a 6-DOF AUV are given; practical aspects of the implementation are also discussed. The experiments have been conducted in the pool of the University of Hawaii using ODIN, an AUV developed at

  • PDF / 787,054 Bytes
  • 10 Pages / 439.37 x 666.142 pts Page_size
  • 15 Downloads / 208 Views

DOWNLOAD

REPORT


Experiments of Dynamic Control of a 6-DOF AUV

5.1 Introduction In this chapter some experimental results on dynamic control of a 6-DOF AUV are given; practical aspects of the implementation are also discussed. The experiments have been conducted in the pool of the University of Hawaii using ODIN, an AUV developed at the Autonomous Systems Laboratory (ASL) [1]. Implemented control law. The control law is briefly rewritten: uv = B †v [K D sv + Φ v (Q, ν, ν˙ a )θˆ v ] ˙ T ˙ a )sv θˆ v = K −1 θ Φ v (Q, ν, ν

(5.1) (5.2)

where B †v is the pseudoinverse of matrix B v (see the Appendix), K θ > O and Φ v is the vehicle regressor. The vectors sv ∈ R6 and sv ∈ R6 are defined as follows sv

   B  ν˜ 1 R I η˜ 1 −1 + Λ+ KD KP = ν˜ 2 ε˜   = ν˜ + Λ + K −1 D K P ˜y, 

sv = ν˜ + Λ ˜y ,

(5.3) (5.4)

T  with η˜ 1 = xd − x yd − y z d − z , ν˜ 1 = ν 1,d − ν 1 , where the subscript d denotes desired values for the relevant variables. The matrix Λ ∈ R6×6 is defined as Λ = blockdiag{λ p I 3 , λo I 3 }, Λ > O. The matrix K P ∈ R6×6 is defined as K P = blockdiag{k p I 3 , ko I 3 }, K P > O. Finally, it is ν a = ν d + Λ ˜y and K D > O.

G. Antonelli, Underwater Robots, Springer Tracts in Advanced Robotics 96, DOI: 10.1007/978-3-319-02877-4_5, © Springer International Publishing Switzerland 2014

117

118

5 Experiments of Dynamic Control of a 6-DOF AUV

Fig. 5.1 ODIN, the AUV used to experimentally test adaptive and fault tolerant control strategies

5.2 Experimental Set-Up ODIN is an autonomous underwater vehicle developed at the Autonomous Systems Laboratory of the University of Hawaii. A picture of the vehicle is shown in Fig. 5.1. It has a near-spherical shape with horizontal diameter of 0.63 m and vertical diameter of 0.61 m, made of anodizied Aluminum (AL 6061-T6). Its dry weight is about 125 kg. It is, thus, slightly positive buoyant. The processor is a Motorola 68040/33 MHz working with VxWorks 5.2 operating system. The power supply is furnished by 24 Lead Gel batteries, 20 for the thrusters and 4 for the CPU, which provide about 2 h of autonomous operations. The actuating system is made of 8 marine propellers built at ASL; they are actuated by brushless motors. Each motor weighs about 1 kg and can provide a maximum thrust force of about 27 N. The sensory system is composed of: a pressure sensor for depth measuring, with an accuracy of 3 cm; 8 sonars for position reconstruction and navigation, each with a range 0.1 ÷ 14.4 m; an Inertial System for attitude and velocity measures.

5.3 Experiments of Dynamic Control Despite the closed environment in which the experiments have been conducted, the pool of the University of Hawaii, it is necessary to take into account the presence of a current as an irrotational, constant disturbance [2]. The modeling aspects of including the current in the dynamic model have been discussed in Sect. 2.4.3. Since the measure of the current is not available in ODIN, it has been taken into account

5.3 Experiments of Dynamic Control

119

as a disturbance τ v,C acting at the force/moment le