Here's a nice view of the undercarriage in action. In California we used spars arranged in an 'H' configuration much like a conventional helicopter, but the first thing you need to know about multicopters like this is that they are far from conventional helicopters. With an engine failure or when struggling with thin air, helicopters are able to perform a running landing, which is why they have upturned skids. Quadcopters land vertically and are not oriented in any direction ~ 'forward' here defined purely by the passenger, who may face any direction independent of the aircraft.
Accordingly an arrangement like this is likely sufficient whilst at the same time being aesthetically pleasing, echoing as it does the cantilevers overhead that support the perimeter of the rotorhead. As a consequence I have made them the same length at 26" apiece not counting the tube inserts, which are domed at the working end and flat at the other.
During the shake-down here I have assembled the different modules viz. landing legs, foot-well, connecting rods, passenger restraint and rotorhead in order to see how they would be assembled in the field. I am unhappy with the way the latter attached to the top of the four columns, which was unduly fiddly and set to be moreso at full-scale.
As a result I make two key decisions: we shall switch to 1" tubing for the uprights and at the same time include four matching holes in the centre-section of the drone, so that it can be conveniently fitted at ground-level and 'flown' to the top for boarding.