Successive convex optimization for transformer encoder model predictive control

Conference item

We propose a data-driven Model Predictive Control (MPC) framework that employs a transformer encoder to generate multi-step predictions. To handle the nonconvex attention mechanism, we derive difference of convex (DC) representations of the transformer encoder components and embed them in a successive convex programming (SCP) iteration. Recursive feasibility and convergence of the SCP iterates are guaranteed, and each iterate yields a solution estimate satisfying the problem constraints. Under mild assumptions, the SCP iteration converges to a locally optimal solution of the MPC problem. The approach is illustrated on a benchmark nonlinear control problem.

Authors: Chen, X; Cannon, M

• Publication status: Accepted
• Peer review status: Peer reviewed
• Acceptance date: 2026-05-14
• Event title: 23rd IFAC World Congress
• Event location: Busan, Republic of Korea
• Event website: http://www.ifac2026.org/
• Event start date: 2026-08-23
• Event end date: 2026-08-28
• Language: English
• Keywords: transformer encoder; DC programming; successive convex programming; data-driven model predictive control