Maintaining receiver’s thermal stresses and corrosion below the material limits are issues that need careful attention in solar thermal towers. Both depend on heliostats’ aiming points over the central receiver and available direct solar radiation at any instant. Since this technology relies on an unavoidable time-changing resource, aiming points need to be properly manipulated to avoid excessive hot spots. This paper proposes a new aiming point strategy based on a multivariable model predictive control (MPC) approach. It shows an alternative approach by introducing an agent-based group behavior over heliostats’ subsets, which makes possible either concentrating or dispersing solar radiation as required by the MPC algorithm. Simulated results indicate that it is feasible to develop a closed-loop control procedure that distributes solar irradiance over the central receiver according to the predefined heat flux limits. The performance of the proposed approach is also compared with the results found in the available literature that uses a different methodology.