Heavy Lift Helicopter Teams

Motivation

There exists a growing demand for heavy lift aerial transportation systems. Most research and development focuses on the creation of larger, more powerful and expensive lifting vehicles, such as Boeing’s JHL-40 (Skyhook). These systems accomplish the task by scaling up existing technology to provide increased payload capability but unfortunately at increased complexity and cost.

As the vehicles become larger, more attention needs to be paid to the structural dynamics. Larger airframes result in slower, significant vibrational modes which couple into the rotation and translation dynamics. The ensuing higher-order system requires study and simulation to guarantee safety and adequate performance.

An alternative approach would be to possibly use smaller, existing unmanned autonomous helicopters working cooperatively to perform the same task. This proposed system would provide a scalable solution that can be tailored to suit any assignment. Using smaller helicopters places more emphasis on the team strategy and control rather than the control of the individual helicopter. It may also be possible to reduce load swings during transport through the use of distributed connections.

Using a distributed solution such as this also allows for improved sensing capabilities. The sensing capability of one agent can be dramatically increased by incorporating other agent information. Therefore, by working together agents can achieve the high precision necessary for control.

Challenges

Lifting a load using helicopters has been studied and possible solutions have been proposed using specialized control architectures, support structures and passive control devices. Although these solutions have been shown to work in practice, they are ultimately limited by the number of helicopters that can be employed.  We believe the real challenge occurs when more helicopters are added.

For helicopters in two dimensions the position of the load can be uniquely determined by the position of the two lifters. However, when a third lifter is added, the problem changes significantly. When the tethers are modelled as inelastic cords, the tethers become holonomic constraints. Therefore, each lifter may only move on a circle given by the load connection point and the length of the tether. Should a lifter move to a point off of the circle then the system changes, see the above figure. With a small perturbation one or two of the tethers can become slack, thus rendering the associated lifters ineffective, possibly causing the full system to become unstable.