A penalty method for optimal control

Optimal control problems with inequality constaints are often solved by penalty function methods. This letter describes a new penalty function approach extending to optimal control the recent work of Fiacco and McCormick in the mathematical programming area. The most important theoretical results are also summarized.