The optimal use of the solar powered automobile

Once a solar car has been built and tested, the question arises as to how to race with it. Since most teams are extremely energy limited for a race, strategy becomes critical if one is to win. In Sunrayce '97, most of the terrain, as well as the data on stop signs and stop lights, was known by the Illinois team many months before the race. We needed to determine at what speed we should be traveling during all parts of the race in order to use energy most efficiently and finish the race in the least amount of time. In what way should one optimally accelerate from a stop? Should we cruise at a constant speed, at a constant power, or in some other way? How should one climb and descend hills? Also, given a weather prediction for tomorrow, how much energy should one aim to consume today? These are some of the questions we answer through our analysis. In the next section, some of the prior work is described, followed by derivations of the equations modeling the car. The following section formulates the optimal control problem when the weather conditions are known, followed by solutions for various scenarios of interest. The next two sections first compare the results with different rule of thumb strategies and then obtain the optimal strategy given probabilistic weather predictions using stochastic dynamic programming. A post-race analysis is then provided.