Using measurable physical input variables, an implementable control algorithm for parallel architecture plug-in and non-plug-in hybrid electric vehicle (PHEV and HEV) powertrains is presented. The control of the electric drive is based on an algebraic mapping of the accelerator pedal position, the battery state-of-charge (SOC), and the vehicle velocity into a motor controller input torque command. This mapping is developed using a sequential linearization control (SLC) methodology. The internal combustion engine (ICE) control uses a modified accelerator pedal to throttle plate angle using an adjustable gain parameter that, in turn, determines the sustained battery SOC. Searches over an admissible control space or the use of pre-defined look-up tables are thus avoided. Actual on-road results for a Ford Explorer with a through-the-road (TTR) hybrid powertrain using this control methodology are presented. In addition, Matlab-based simulation results for PHEVs and HEVs using this single control algorithm are presented employing a model of the Explorer. These results show the gasoline consumption and SOC over 48 repeated cycles of three common driving schedules. Further studies of acceleration performance illustrate the trade-off of minimizing gasoline consumption vs. higher performance using more gasoline. All-electric range (AER) vs. battery capacity is also presented as well as the fuel consumption and final SOC for two PHEV control strategies.