Approximate Dynamic Programming Applied to a Four Quadrant Transducer Series-Parallel Hybrid Electric Bus

Paper in proceeding, 2009

This paper presents an Approximate Dynamic Programming scheme that, based on the assumption of perfectly known future driving, efficiently solves the optimal power split in a Series-Parallel hybrid powertrain. The scheme combines two ideas to reduce the computational load of the dynamic programming. First, the cost-to-go function is approximated using piecewise linear functions on a sparse grid. Secondly, by using precalculated maps, the iterations performed in the dynamic programming are reduced to simple table lookups and linear interpolation. The approximation scheme is simulated on a Four Quadrant Transducer Series-Parallel Hybrid Electric Bus. In the simulations the approximation scheme is able to find a sufficiently good approximation of the optimal control trajectory.