We would like to describe an alternative approach to the development of a ROV system, which has lead to a simplified homogeneous design We feel that this approach has significantly reduced the time and effort required to move from the prototyping to production phase of our ROV system
The approach we are proposing is based on techniques of problem analysis developed in Computer Science The paper will describe the analytical technique and its application to the design of a particular ROV system, described in terms of an information engineering problem
The design of computer systems is becoming a refined art Recent advances made in computational science mean that it is now possible to prove mathematically that a system is correct, i e the system is an accurate implementation of its description, that is, it does what you think it is supposed to do, no more no less This has been a very significant result in the area of computational mathematics, but unfortunately is still limited a restricted class of problems
However, these proof techniques may be used more generally to formulate a heuristic which can be applied to more generalised problems, with the intent of leading to more optimal solutions. In computer science, a heuristic is method of solving a problem derived from trial and error So in this case the heuristic is a derivative of these techniques The essential difference between the heuristic and proof, or algorithm is that a solution is not guaranteed at the end of the day.
The design of an ROV may on the surface appear to be a simple task, however during the average design process it soon becomes apparent that to achieve a satisfactory result the balancing of compromise becomes a most prominent factor, playing off one factor against another hoping to obtain the best end product
The main objective of the technique to be described, is to attempt to identify difficulties in a problem or project during the definition phase before it is too late, that is, before you are committed into hardware Otherwise it is likely that the result will be a series of special case pragmatic solutions, each maybe effective in themselves, but without an overall homogeneous structure Lack of homogeneous structure means that the solution is rigid
As rigidity is the converse of flexibility, this implies that if the solution is a series of special cases, then it will be much more difficult to modify such a solution
When the definition phase has been completed, a clear view of the solution will have emerged, and consequently at this stage clearer views of costings, timings and so on may be achieved From this point the level of confidence in the ability to attain a successful completion of the project rises dramatically