Several spatially distributed tasks can be completed more efficiently by using a combination of better sensing, prediction, and planning. When an agent needs to regularly move to serve different locations, it is necessary to find a compromise solution between the travel cost and the required quality of service. This problem is present in several contexts, namely waste collection, maintenance, and surveillance. This problem is very complex since it includes sensor placement and routing problems that, due to the uncertainty and combinatorial decision making, cannot be solved optimally in an efficient way. In this work, we develop a new efficient algorithm for sensor placement. It takes into account scheduling and route planning to serve distributed locations, considering not only the value of the information provided by each sensor but also the impact on travel efficiency and quality of service. We present examples of sensor placement and route planning in a waste collection environment.
展开▼
