Factoring Communications and Situational Awareness in Operational Models of Dismounted Combat
Abstract
Defence funds dedicated to dismounted soldier systems are finite, and must be divided among multiple components. Deciding on the right mix can be difficult some technologies improve lethality and protection, others improve SA. At the same time, these technologies might increase cognitive and physical load. In this report we present a way to perform comparisons across this apparent divide, and find the optimal mix of technologies. We present a mathematical combat model that considers the joint effects of situational awareness, lethality, and protection equipment in terms of expected lives saved. The model can therefore be used to design an optimal dismounted soldier system, one that will save the most lives. Model Our approach relies on representing the decision maker as an optimal one, at all times. That decision maker however must make decisions under uncertainty, and time constraints. As cognitive burden increases, several changes can occur in the model: the time between decisions might increase, the amount of information considered in each decision can decrease, or the planning horizon might be shortened, resulting in more myopic decisions. Each of these levers in the model gives the flexibility to represent a degradation of decision-making, and SA, while still assuming that the commander is making the best decision possible, but under difficult constraints. Technically, our model is based on two pillars. First, combat is modelled as a Continuous-Time Markov Chain (CTMC). Second, the commander is modelled as a decision-maker in a Partially-Observed Markov Decision Process (POMDP). POMDPs are sequential decision problems that a resolved by dynamic programming. They are difficult to solve because, contrary to fully observable Markov Decision Process (MDPs), some of the state variables are hidden. Fortunately, advanced computational methods have been developed to solve them.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2020
- Accession Number
- AD1183704
Entities
People
- Andrew Coutts
- Christopher J. Nicholls
- Jerome Levesque
- K. T. Smith
- Maurice Vanbeurden
- Nicholas H. Stanbridge
- Roy Benda
Organizations
- NATO Science and Technology Organization