Abstract
This paper is devoted to modelling and control algorithms for a slung load transportation system composed of one or multiple helicopters, where the load is coupled by the means of flexible ropes (see Fig. 1). The coupled helicopter system overcomes the payload limitation of a single small size helicopter, while keeping most of its advantages: small costs of operation, low maintenance costs and increased safeness. Therefore, the system can be utilised whenever the use of full size helicopters is impossible, too expensive or prohibited by law. We focus on the deployment and repairing of distributed sensor networks, using a transportation system based on multiple small size helicopters. A possible real world application is the deployment of fire fighting equipment, where space limitation of the fire trucks prohibits the application of bigger UAVs and using full size helicopters is too dangerous. The problem of load transportation using one or two full size helicopters (twin lift helicopter system), connected to the load by means of flexible ropes, has been discussed in the aerospace research community at least since 1960. We have shown in our previous work that there is a fundamental difference in the mathematical description between small and full size helicopters. Therefore, also the control design for the case of small size helicopters needs to be different. To our knowledge, the control of a slung load transportation system composed of multiple small size helicopters has not been studied until now. In this paper, the complete mechanical setup of the slung load transportation system based on one or more small size helicopters is presented. This includes a short description of the used UAVs, the additionally required sensors, and how the load is mounted. A model of one/multiple helicopters transporting a load is introduced. This model is used in a simplified form for the controller design and in full form for simulation. The controller for one and two helicopters, which is based on a state feedback controller, as well as the controller for three and more helicopters, which is based on a non linear controller, are explained in detail. Both controllers utilise an underlying non-linear orientation controller. We propose a feedback loop, based on forces measured in the ropes, to compensate for the influence of the rope. The controllers were tested in simulation and in real flight experiments. The world wide first flight experiment with three coupled helicopters was successfully conducted at the end of 2007.
Task planning and control for a multi-UAV system: architecture and algorithms