No abstract

No abstract

This paper advocates the Gaussian belief propagation solver for factor graphs in the case of gas distribution mapping to support an olfactory sensing robot. The local message passing of belief propagation moves away from the standard Cholesky decomposition technique, which avoids solving the entire factor graph at once and allows for only areas of interest to be updated more effectively. Implementing a local solver means that iterative updates to the distribution map can be achieved orders of magnitude quicker than conventional direct solvers which scale computationally to the size of the map. After defining the belief propagation algorithm for gas mapping, several state of the art message scheduling algorithms are tested in simulation against the standard Cholesky solver for their ability to converge to the exact solution. Testing shows that under the wildfire scheduling method for a large urban scenario, that distribution maps can be iterated at least 10 times faster whilst still maintaining exact solutions. This move to an efficient local framework allows future works to consider 3D mapping, predictive utility and multi-robot distributed mapping.

We utilise collaborative path-finding to improve efficiency of smart parking systems and therefore reduce traffic congestion in metropolitan environments, while increasing efficiency and profitability of parking garages. A significant portion of traffic in urban areas is accounted for by drivers searching for an available parking space. Many cities have adopted a parking guidance and information system to try to alleviate this traffic congestion. Typically these systems entail informing the driver of the whereabouts of an available space, reserving that space for the specific driver, and providing directions to reach the destination. Little or no account is taken of how much congestion will be caused by multiple drivers being directed to the same car-park concurrently. We introduce the concept of collaborative path-finding to the problem. We simulate a smart parking system for an urban environment, and show that a novel approach to collaboratively planning paths for multiple agents can lead to reduced traffic congestion on routes toward busy parking areas, while reducing the amount of time when parking spaces are vacant, thereby increasing the revenue earned.

No abstract