Solution space of ODEs

In the context set here for ODEs, solutions can be understood as integral curves of the vector field A. Using terminology of moving frames of Olver, we can consider the local group of transformations given by the vector field A, and the cross section K={(x,u(m1))Jm1:x=0}. The fundamental invariants of the solutions are nothing else than the "inicial conditions": c0,c2,,cm1, which constitute a "space" (not necessarily a manifold, maybe an orbifold?) of dimension m.

In the case of a homogeneous linear ODE, the solution space constitute a vector space. It is easy to check that the space of fundamental invariants has also the structure of a vector space, an the dimension is m, of course.
Pasted image 20211110075411.png

Related with the independence of solutions of a linear ODE is the Wronskian.