| In this work, multi-objective decision making techniques are utilized to address the problem of controlling an autonomous physical agent in real-world environments that present simultaneous competing navigation and sensing objectives. The long-term objective of this work is to dramatically advance the capability of an autonomous agent to act effectively in a physical environment. In short, the agent must make good decisions with good information. In practice, the quality of the information is tied not only to the quality of the sensors and sensor algorithms, but also to the manner in which the agent positions and utilizes those sensors. Doing the latter effectively, is often in contradiction or competition with agent actions that more directly relate to mission objectives. The common thread addressed in this work is the ability to make good control decisions that balance the multiple needs of sensing and acting in real environments. |
| In the first stage of this collaboration, the interval programming (IvP) algorithms have been integrated into the MOOS software libraries. MOOS offers an ideal way to embed the multi-objective optimisation framework resulting in a product that has allowed for deployment on several vehicle platforms land, surface and sub-sea). IvP is a generic optimizer and so a plethora of real-world applications are relevant following successful integration with MOOS. However, of particular interest to both parties is the use of IvP to determine optimal sensing actions given a suite of various sensors deployed over a fleet of mobile robots. We will further investigate how IvP can be used to produce sensible vehicle trajectories and sensing actions (e.g. gaze of a camera ) while imposing frequently conflicting mission criteria . Consider the complexity of maximising energy efficiency, rate of information gain, navigation precision/stability while operating in non-trivial environment - one populated with exogenous moving entities such as people or vehicles. |
| The IvP libraries are available to other ONR or US Gov't funded projects upon email request. I intend to put this code under a GPL or similar license as soon as possible pending navy approval. They are, in short, an implementation of the ideas described in the papers below, (in the public domain), after many years of fine-tuning, simplifying and testing. |