Research

In this project, various techniques will be investigated for autonomous driving, including the use of smart markers to transmit and receive critical navigation information through and around obstacles or minefields. Successful development of this technology could be applicable not only to manned vehicles and dismounted forces but also to the path planning and path following of unmanned vehicles.
 

This research is expected to develop autonomous multi-robot algorithms for inherently cooperative tasks in presence of different levels of communication between robots. In particular, learning algorithms in the domain of cooperative multi-robot observation of multiple moving targets will be investigated. These innovative techniques allow robots to build up memories of their experiences in the environment, evaluate the utility of alternative cooperative actions, and then select actions to take that increase the likelihood that the desired global team goals will be achieved through the individual robot decisions.
Specific technologies include intelligent driving decision aids, the application of semiautonomous driving technology, and automated route planning, all of which are pertinent to both manned and unmanned vehicles.

The Guidance, Navigation and Control (GNC) requirements for the robots composing the multi-robot system must similarly take the scenario elements into account. The definition and subsequent fulfilment of the GNC requirements will be consequently justified in light of the necessities of the system with respect to the scenario definitions. This will be done in a hierarchical manner within a top-down approach. Those assumptions which are necessary in order to maintain a workable and streamlined set of requirements will be pointed and discussed.

Also, the component of each robot's Guidance, Navigation and Control (GNC) system takes into account the estimation and mapping of its surrounding environment. The robot receives an estimate of its own internal state, which is combined with external sensor inputs from the laser, ultrasonic and camera sensors to arrive at a 3D estimation of the surrounding terrain and other active dynamic elements (robots, persons, vehicles, etc.). Furthermore, in the absence of a reliable state estimate, the techniques of Simultaneous Localisation and Mapping (SLAM) will be used to improve this estimate based upon matching principles between the estimated and prior information of the terrain.

Thus, three complex, difficult and still relatively immature technologies will be applied in the field of robot navigation: 3D terrain estimation, dynamic object estimation and SLAM.