The paper presents models and algorithms for planning and simulating the movement of multiple objects in knowledge-based military DSS and CGF systems. The case study concerns the concept, decision-making model, and control associated with a decision-making automaton for marches at the battalion level. The automaton implements two main processes: planning the march decision and directly controlling the course (simulation) of the march. Selected theoretical and practical aspects of modeling and optimizing the march planning process, simulation, and control of its progress are discussed. The implementation method of the automaton and selected simulation results in the interactive operational training support system Złocień are described.INTRODUCTION
One of the most significant transport problems encountered in the literature are problems related to movement planning. These problems are important not only in military applications but also in: computer networks, mobile robots, evacuation systems, car navigation systems, and computer games
[8]. In military applications, these problems are encountered both in battlefield simulators (determining routes before the start of operations simulations, as well as during their duration), as well as in decision support systems that support automated C4ISR (Command, Control, Communications, Computers, Intelligence, Surveillance and Reconnaissance) command systems [3], [5], [7], [9], [10], [14], [15].
The described systems should be knowledge-based, and decision support and simulation models and algorithms should utilize this knowledge [4], [13]. Knowledge is information processed based on certain rules, derived from various types of thematic databases (about terrain, tactical action regulations, weapons and military equipment, military structures, etc.) and includes elements such as: passability of indicated terrain fragments, patterns of decision-making situations, patterns of action variants for different types of operations, etc. Knowledge-based simulation, i.e., using artificial intelligence models and methods, has been developed since the late seventies and early eighties of the last century, and the pioneer of this research was the RAND Corporation in the USA. (…)
The article contains 37,575 characters.
Source: Logistyka Magazine 4/2011