Stataflow error - does not have continuous states

I am doing a stateflow using MATLAB simulink and the attached errors are showing. Can someone please help me solve this? Attached please also find the stateflow that I am using. Thnaks in advance.

[Walter Roberson]

Does [PROX == 0] indicate a conditional flow or does it indicate an assignment of 0 to PROX ?

[John Smith]

Thanks for the reply. I didn't really understand what you mean, but PROX is an event set by a manual switch (either 1 or 0 - constant box)

[Walter Roberson]

Your chart looks like when you are IDLE and PROX tests as 0, to go to Counter, with no behavior defined at IDLE for the case where PROX is non-zero. Then once you are at Counter (implying that PROX is 0), then it appears that if PROX tests to be 1 then Calculation is gone to, with no behavior defined from Counter if PROX is not 0. But since PROX cannot be both 0 and 1, if you get to Counter because PROX is 0, you can't get to Calculation because PROX is not 1; you also cannot get to Calculation any other way.

That's how I would read the state diagram.

[John Smith]

I removed the NORMAL_MODE and EMERGENCY_MODE parent states and the same thing is showing. I am attaching the schematic. Any kind of help would be highly appreciated.

[Walter Roberson]

What is the current set of messages?

[Walter Roberson]

Also the current Chart

[John Smith]

The screenshots I posted before are of the original schematic. Below are the schematic, states and Stateflow debugging I am having when I remove some on the states for testing. (p.s. sorry for my poor English)

[John Smith]

Hi again, I realized that the pulse generator was causing the problem. When I replaced the generator with a manual switch, everything worked well. What am I doing wrong please? I connected the generator to a mux since it is a shared event. I set the period to 2sec with 50% cycle. What is causing this please?

[Walter Roberson]

The bit about Continuous States was never a problem. It was a note telling you that you were working with a discrete time system instead of a continuous time system. ode solvers are only useful for continuous time systems. I imagine that continuous time could have been forced by selecting a different kind of signal, but I am not familiar with how those operate.

The only bit I was concerned about was the part about some of the states being only conditionally reachable, leading to potential instability. And that appears to have been solved.