The Need

Hybrid and electrified vehicles face increasing pressure to simultaneously reduce fuel consumption and tailpipe emissions, particularly during transient operating conditions such as cold start. Conventional rule-based or static control strategies struggle to optimally manage the tradeoffs among engine operation, electric motor usage, battery state of charge, and exhaust aftertreatment thermal requirements. This results in suboptimal fuel economy, delayed catalyst light-off, higher cold-start emissions, and extensive calibration effort across drive cycles, vehicle platforms, and operating conditions.

The Technology

OSU engineers have developed an AI-enhanced, nonlinear Model Predictive Control (MPC) framework that coordinates power split and thermal management in hybrid and mild-hybrid powertrains. By leveraging route preview, predicted driver behavior, and real-time vehicle states, the controller dynamically optimizes engine and motor torque commands while managing battery state of charge and exhaust catalyst heating. The approach unifies energy management and aftertreatment thermal control within a single predictive optimization framework, enabling fuel- and emission-aware decision-making across an entire trip.

Benefits/Advantages

• Improved fuel economy: Predictive optimization reduces total fuel consumption over real-world drive cycles
• Faster catalyst light-off: Coordinated electrical heating and engine load management reduce cold-start emissions
• Reduced calibration burden: Adaptive, model-based control minimizes manual tuning across platforms
• Scalable and modular: Compatible with existing ECMS and deployable across multiple hybrid architectures