essay代写,代写assignment,paper代写,代写留学作业,英国作业

导航切换

QQ:
153688106

二维码

当前位置:主页 > 代写coursework >

代写英国coursework-An Interactive

浏览: 日期:2020-01-13

An Interactive Planner for Tank Squadron Assaults Abstract A planner has been developed for planning assault tasks for a group of tanks. This planner is intended for use in training novice commanders and for planning the actions of a group in simulations. Because the notion of a correct solution varies widely between subject matter experts and may depend on information unavailable to the planner an interactive approach has been taken. The resulting system carries out analysis of the terrain to gauge its suitability for the assault and applies a set of constraints to identify a solution appropriate to the terrain. At any stage the user may override the automated planning and either select their own preferred solution or alter the planning parameters to produce solutions more acceptable to their needs. The resulting system can therefore be used as a fully automated planner, a user assistant, or simply as an input device for the user’s plan. To achieve this it was necessary to provide graphical output of the results of each planning stage and to model the progress of planning and the level of authority delegated by the user for each step. 1. Introduction Computer Generated Forces (CGF) are used in battlefield simulations for training exercises, as well as for operational analysis. The use of CGF reduces the number of human operators required and so reduces training costs, hence there is a desire to increase the amount of automation. This requires a computer to be able to plan complex activities. However, users of planning systems tend to prefer those that allow them to influence the planning over systems which operate as a black-box [1]. Therefore our work has been to investigate how an existing fully automated planner can be modified to produce an interactive, or mixed initiative, planner. The Broad Agents [2] system developed between 1994 and 1998 at DERA Malvern in POPLOG contains an automated planner for tank squadron assaults [3]. It incorporates terrain reasoning in order to produce a plan to attack a specifie