Fig 1.
Part of hardware and software architecture of the flight control system.
The figure shows parts of the organizational structure of a flight control system, and describes the data interaction among module.
Fig 2.
We can use MARTE time stereotypes to define and describe clock class.
Fig 3.
The software logic structure of flight control system.
Using RtUnited and PpUnit, we build part of the software logic modle of flight control system.
Fig 4.
State chart of flight control system.
State chart can be used to describe dynamic behavior triggered by event.
Fig 5.
FG module described in Object-Z class.
Object-Z is a formal language, it can be used to model the static model accurately.
Fig 6.
PTA expression in Object-Z format.
Dynamic state transition can be described by the state schema in Object-Z.
Fig 7.
Scenario-oriented requirement for flight control software.
MARTE time model is used to describe the requirement. With timing constraints.
Fig 8.
The expected time of each status within T unit time described in logarithm form.
(a) The expected time within 80ms. (b) The expected time within 60m, (c) The expected time within 24h. (a)Within 80ms, the probability of system shutdown is smaller. (b) As time goes on, the danger status goes up. The UAV is fit to carry out a short-term mission. (c)The reliability of hardware determines UAV whether can perform tasks for a long time.
Table 1.
The Expected time in states “danger” and “up”.
Fig 9.
The reliability probability of each processor.
With the increasing K value, the reliability probability became stable.