STLCCP: Efficient Convex Optimization-Based Framework for Signal Temporal Logic Specifications
STL is expressive, but its native mixed-integer encoding does not scale to long horizons. STLCCP recasts the synthesis problem by exploiting three structural properties of STL — monotonicity of the robustness function, its hierarchical tree structure, and the correspondence between conjunction/disjunction and convexity/concavity. The result is a difference-of-convex program solved as a sequence of convex quadratic programs, augmented by a tree-weighted penalty CCP and a tailored mellowmin smoothing. STL becomes a computationally viable engine for control over long horizons.