How can AI algorithms based on learning and based on reasoning effectively be combined to ensure real time Situational Awareness (SA) while executing a robust mission plan for combat unmanned ground vehicles (CUGVs) to enhance their capabilities, including the ability to disturb/degrade/destroy enemy fire support attacks?

Problem Context

In this project we want to obtain Situational Awareness (SA) and Situational Understanding (SU) through a combination of learning based detection methods and reasoning algorithms. The context of the work is that of tactical military operations, and the focus of the current (simulated) application is on ground-based platforms. The reference behavior for the UxVs is defined by military doctrine. The conception of a mission plan for the team of UAVs and UGVs is a challenging task, because the plan has to take, amongst others, different enemy course of actions into account, while remaining within the constraints of the mission and agreed upon plan before mission execution.


The goal is to construct a comprehensive mission plan for multiple armed UGVs to conduct an interdiction and strike mission, have the UGVs carry out that mission plan, and adjust the details of the tasks in the plan as the situation changes. A reasoning approach for autonomous unmanned vehicle (UxV) mission execution is presented, taking both mission planning and mission execution into account, treating these as separate but closely interdependent stages.


The implemented approach to both planning and execution of a mission of unmanned vehicles was brought a step forward, with a focus on the execution, and the concept has been proven to be a feasible approach. In addition, significant steps were made in creating an approach to searching and identifying enemy vehicles for a UAV with a camera, taking into account parameters such as the terrain, knowledge of the enemy doctrine, and characteristics of the camera characteristics.


  • Jan de Gier, Research scientist autonomous systems, TNO, e-mail: