Flight Log: Testing APM 2.0 Autopilot Mode Successful (Take off and RTL)

In today’s test, I patiently waited for the GPS to lock under nice weather condition.
No more initial manual flight, in which once the GPS locked, the quadcopter were brought to the middle of the field and binded to my Devo 7 transmitter. The waypoint definition can be seen below, in which I would like the quadcopter to fly up to 6 meters, and initiate return to launch (RTL).

quadcopter UMT

My student and I went to a safe distance. I initiated 5% throttle, and flipped the autopilot switch. The quadcopter steadily hovered up to approximately 6 meters (I could not confirm this because I didn’t install the telemetry, the APM 2.0 did not like it). The quad was pushed by wind and it tried it best to land by searching for the launching coordinate to no avail. Before it hits the building, I regained manual control and landed it safely.

On the second try, the quad steadily flies up to approximately 6 meters, stayed for 3 seconds and landed safely on the initial coordinate. This means victory! To confirm that the quadcopter landed not because of the weak battery, I regained manual control and flied the machine. The quadcopter flied successfully indicating that the autopilot mode was working just fine.

The question is, should I reduce the waypoint radius? The current one is 30 meters. Should the waypoint radius reduce the hover of the quad and make it easier for it to land?