A new approach to the self-dual Yang-Mills equations (Q801588): Difference between revisions

The new method established by \textit{M. Sato} [Random systems and dynamical systems, Proc. Symp., Kyoto 1981, RIMS Kokyuroku 439, 30-46 (1981, Zbl 0507.58029)] in the theory of two dimensional completely integrable systems is considered in the framework of the four dimensional self-dual Yang-Mills (SDYM) equations. The basic idea is to reduce the well known commutator representation of (SDYM) to the equations on the infinite dimensional Grassmann manifold. The complexified (SDYM) with the structure group GL(r,\({\mathbb{C}})\) are written in the form: (1) \(\partial_ yA_{\bar z}-\partial_{\bar z}A_{\bar y}+[A_{\bar y},A_{\bar z}]=0\), \(\partial_ yA_{\bar y}+\partial_{\bar z}A_{\bar z}=0\), where \(\partial_ u=\partial /\partial u\), \(u=y,\bar y,z,\bar z\) and \(A_ u\) denotes the \(g\ell (r,{\mathbb{C}})\) valued functions. The author introduces, following Sato, an \(\infty \times \infty\) matrix \(\zeta =(\zeta_{ij})_{i\in {\mathbb{Z}},j<0}\) consisting of \(g\ell (r,{\mathbb{C}})\) valued unknown functions \(\zeta_{ij}\) and considers the equations: (2) \(-\partial_ y\zeta_{i+1,j}+\partial_{\bar z}\zeta_{ij}+\zeta_{i,-1}\partial_ y\zeta_{0j}=0\), \(\partial_ z\zeta_{i+1,j}+\partial_ y\zeta_{ij}-\zeta_{i,-1}\partial_ z\zeta_{0j}=0\), \(\zeta_{i+1,j}=\zeta_{i,j-1}+\zeta_{i,- 1}\zeta_{0j}\) for \(i\in {\mathbb{Z}}\), \(j<0\), \(\zeta_{ij}=\delta_{ij}I_ r\), \(i,j<0\). Here \(I_ r\) denotes the \(r\times r\) unit matrix. The author explicitly constructs the solutions of the system (2) and shows that the following equations: \(A_{\bar y}=\partial_ z\zeta_{0,-1}\), \(A_{\bar z}=-\partial_ y\zeta_{0,- 1}\) define a correspondence of solutions between (1) and (2). The author compares this method with the inverse scattering method.