Added section on singular value decomposition

core/tutorial/doc/06_algebra.md

@@ -0,0 +1,33 @@
+# Tutorial 06 - Linear Algebra
+
+## Singular value decomposition
+_Note: Currently, calculating the SVD of a matrix requires LinAL_
+
+Use the _SVD_ class for fast singular value decomposition.
+The calculation is performed at initialization and the resulting
+matrices are available from getter functions.
+
+See this short example:
+
+```c++
+// Create a matrix
+NICE::MatrixT<double> my_matrix(3,3,0);
+my_matrix.addIdentity(1);
+my_matrix(0,2) = 5;
+my_matrix(1,0) = 3;
+
+// Calculate the SVD
+NICE::SVD svd(my_matrix);
+
+// Store the results
+NICE::MatrixT<double> result_vt = svd.getVt();
+NICE::MatrixT<double> result_u = svd.getU();
+
+// This returns the singular values in a diagonal matrix
+// This diagonal matrix is allocated only if you call this function.
+NICE::MatrixT<double> result_s_matrix = svd.getS();
+
+// This returns the singular values as a vector. Use this function for
+// increased efficiency
+NICE::VectorT<double> result_s_vector = svd.getSingularValues();
+```