Abstract
The construction of a process plan can be thought of as an exercise in logic programming. Given the facts of the situation and the rules of the conditions governing the processes and given the statement of a goal state, a logic program will attempt to prove that goal state. The series of rules and facts used for the proof will be the process plan. When using a logic programming language such as Prolog, if the problem is soluble, a proof will be found. After the facts, rules and goal are stated, the language itself invokes the computation processes needed to find the proof. This approach is especially useful when having to rapidly and flexibly develop new plans. However, if one tries to solve the problem using a simple theorem-proving approach, then the time and memory required to generate the process plan are excessive. A solution to a realistic problem requires attention to the control of the logic program. Control is achieved by using procedures, based on various possible heuristic techniques, to save time and memory. The paper includes details of novel procedures to implement various heuristic methods in Prolog. Various examples from the world of robotics are shown. Sequential process planning using regular Prolog, and planning for parallel and sequential communicating processes as generated by Concurrent Prolog, are demonstrated.
Original language | English |
---|---|
Pages (from-to) | 1-10 |
Number of pages | 10 |
Journal | Robotics and Computer-Integrated Manufacturing |
Volume | 5 |
Issue number | 1 |
DOIs | |
State | Published - 1 Jan 1989 |
ASJC Scopus subject areas
- Control and Systems Engineering
- Software
- General Mathematics
- Computer Science Applications
- Industrial and Manufacturing Engineering