Rigid Body Attitude Estimation- An Overview and Comparative Stud.pdf

# Diag1 1 det u 39 v v diag1 1 det v 310 s diag s 1 s 2

• 153

This preview shows page 29 - 31 out of 153 pages.

[diag(1 , 1 , det U )] , (3.9) V + = V [diag(1 , 1 , det V )] , (3.10) S 0 = diag( s 1 , s 2 , s 3 (det U )(det V )) , (3.11) where det denotes the determinant of a matrix and (det U )(det V ) = ± 1. Then, the matrix B is decomposed into the following form B = U + S 0 V T + , (3.12) and the optimal matrix R opt , which minimizes the cost function (3.1), is found to be R opt = U + V T + = U [diag(1 , 1 , (det U )(det V ))] V T . (3.13) Another version of this method, known as Fast Optimal Attitude Matrix (FOAM) [Markley, 1993], uses the properties of the matrix B to rewrite the optimal rotation matrix (3.13) as R opt = [( κ + k B k 2 ) B + λ adj B T - BB T B ] /ξ, (3.14) where adj denotes the adjoint matrix and k B k 2 = s 2 1 + s 2 2 + s 2 3 , (3.15) and the scalar coe ffi cients κ , λ , and ξ are defined as κ = s 2 s 3 + s 3 s 2 + s 1 s 2 , λ = s 1 + s 2 + s 3 , ξ = ( s 2 + s 3 )( s 3 + s 1 )( s 1 + s 2 ) . (3.16) Since the values of these coe ffi cients depend on the SVD, FOAM takes advantage of an iterative computation strategy to avoid finding s 1 , s 2 , s 3 and instead, directly computing the

Subscribe to view the full document.

C hapter 3. S tatic A ttitude D etermination 22 three scalar coe ffi cients. The coe ffi cients κ and ξ can be expressed in terms of λ and B as κ = 1 2 ( λ 2 - k B k 2 ) , ξ = κλ - det B . (3.17) Using (3.14) and the fact that λ = tr( R opt B T ), λ can be found by solving the following equation ( λ 2 - k B k 2 ) 2 - 8 λ det B - 4 k adj B k 2 = 0 . (3.18) Once this equation is recursively solved to find λ , all the other scalar coe ffi cients can be computed. These will determine the optimal rotation matrix from (3.14). In comparison to other methods. the FOAM algorithm is significantly higher in speed and is shown to be the most robust algorithm among the other deterministic attitude esti- mation methods. It also does not have problems in dealing with the special case of a 180 degrees rotation [Markley and Mortari, 2000], [Markley and Mortari, 1999]. The SVD and FOAM do not adopt quaternion parameterization and work entirely with a rotation ma- trix. This enables them to work without the requirement of computing eigenvalues and eigenvectors and save some computational time. 3.4 Q-Method and the QUEST Since the four component quaternion representation and the rotation matrix are related to each other by simple relations, it can be shown that a search for an optimal matrix R opt in Wahba’s problem leads to the computation of an optimal quaternion corresponding to that rotation matrix [Keat, 1977]. The method, known in literature as the Q-method, simplifies the previous optimization techniques by using the 4 × 1 quaternion vector instead of 3 × 3 rotation matrix. Given the observation pairs of ( ˆ V i , ˆ W i ) and the positive coe ffi cients a i , let us define the following 3 × 3 matrix, 3 × 1 vector z and scalar σ as S : = B + B T = N X i = 1 a i ˆ W i ˆ V T i + ˆ V i ˆ W T i , (3.19)
C hapter 3. S tatic A ttitude D etermination 23 Z : = N X i = 1 a i ˆ W i × ˆ V i , (3.20) σ : = tr( B ) = N X i = 1 a i ˆ V T i ˆ W i . (3.21) Defining the 4 × 4 symmetric matrix K as K = S - σ I 3 Z Z T σ , (3.22) results in (3.1) to be written into the quadratic quaternion function 1 - L ( R ) = g ( Q ) = Q T KQ . (3.23) It is then clear that the minimization of L ( R ) is equivalent to finding the maximum value of
You've reached the end of this preview.

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern