The first step in solving a problem with the complexity of this competition is a quick, and sometimes dirty, analysis. This is also refereed to as a back-of-the envelope analysis, a SWAG - simple wild ass guess, or a look from 30,000 feet.
The purpose of a 30,000 foot analysis is to scope out the effort that is going to be required and identify the major problems that need to be overcome. Each of the major problems is then subjected to a similar analysis. This is repeated until the scope of the problem is understood.
Once this analysis is completed you have a reasonable idea of the feasibility of reaching a solution and how many resources - time, materials, manpower - it is going to require.
I started the analysis by looking at the time constraint of 2 hours, the speed limitation of 2 meters/sec, and the area of 80,000 square meters. Even before doing an analysis my reaction was "That is an awful big area to search!". Little did I know...
The Search Analysis addresses searching the 80,000 square meter "roving area", as it is called in the rules, in the period of two hours and with the speed limitation of 2 meters / second. One interesting observation is that with 10 possible samples a sample must be found every 12 minutes. In Level 1, the pre-cached sample must be located, in a smaller, known, general area, and returned within 30 minutes.
Once the basic problem of the search is analyzed other questions needed to addressed. As the rovers move through the park how do they know where they are? How do they know when a sample is nearby? When approaching the starting platform how do they know if it is the front or the back? Until these, and other questions, are addressed the analysis is incomplete. We are still working at 30,000 feet but if there are no feasible answers to these questions then the challenge itself may be impossible.
The analysis of these issues is covered in two sections. The first, Sensor Analysis, discusses how the rovers know what is around them. The second, Moving Around the Park, and its associated subpage discusses the the navigation and localization problems.
The unique requirements of Phase 1 and Phase 2 are also examined, including some alternatives to the approaches outlined in the earlier analysis. Considering alternatives usually leads to a better understanding of the first analysis, may provide a feasible approach, and serves as a check on the other work.