While loop does weird things with mav commands

Try this code

int main(){
	printf("Moving Forward\n");
	mav(1,500);
	mav(3,500);
	while(digital(15) == 0) msleep(15);
 
	ao();
	printf("Object Encountered - All STOP\n");
 
	msleep(2000); //give it a little rest before backing up
 
	printf("Moving Backward\n");
	mrp(1, 500, -4400);
	mrp(3, 500, -4400);
	bmd(1); bmd(3);
	ao();
 
	printf("Program Complete\n");
 
	return 0;
}

It should work better without calling mav constantly(also you misspelled one of the mav as mmav). You can move the printf if you'd rather have it in the loop.

interesting - didn't think calling the mav command continuously would be a problem -- used to work this way on the old cbc units just fine -- is there something different about the KIPR links?

thanks

I'm not sure if it will cause a problem, but it's best to avoid if you can.

I've also had issues with the mav command. It seems broken.

I haven't worked with a robot for almost a year now, but it didn't seem broken to me, just buggy when it comes to precision(and low velocities)

My experience is that both mav and mrp commands work poorly at speed/velocity of less than 600 in the Link OS. I also had problems of no response if a while loop doesn't have a 10-15 msec msleep delay.
I never put these commands inside a while loop.
Did you know that they do run at speeds to 1500?

I am having a similar problem with running mav() inside a while - I am trying to get a feel for the movement distances and build up some data.

But the motors seem to start randomly inside this loop.

Please see the code below:
int main()

{
//set up six tests
// by having the extra buttons available
extra_buttons_show();
int loop_count = 0;
// go straight
// each test will start with a one second wait after the button is pressed
while (1)
{
//press one of the 6 buttons - s_time is for sleep
int button_pressed = 0;
int s_time = 0;
loop_count++;
printf("loop number is %d \n",loop_count);
ao();

if (a_button()==1) button_pressed =1;
if (b_button()==1) button_pressed =2;
if (c_button()==1) button_pressed =3;
if (x_button()==1) button_pressed =4;
if (y_button()==1) button_pressed =5;
if (z_button()==1) button_pressed =6;

printf("button pressed is number %d \n", button_pressed);
s_time = button_pressed *500;

// after button press, wait one second before beginning movement
msleep(1000);
ao();

printf("go straight %d \n", s_time);
mav(1,1500);
mav(3,1500);
msleep(s_time);
ao();
// freeze(1);
// freeze(3);
}
printf("did this work?\n");
return 0;
}

Any thoughts?

JK

I think what you need is a while loop to sample the buttons on the condition '!button_pressed' that includes each if statement.
As soon as one button is pressed the code will continue with the value selected and run the motors for 1/2 to 3 sec. That will keep looping in while(1) stopping after the loop count and waiting for the next button to be pressed, never printing the last statement.

I've been getting around the wonkiness of 'mav' by using 'motor' command instead -- for whatever reason with the latest version of KISS - mav and while loops are not consistent --- especially when trying to also use an ET sensor to dynamically adjust the speed based on distance

I forgot to say that if you wait for a "button_pressed", after using it to set the run time you need to reset it to '0' so the loop will wait for the next press.
I also want to say that I tried many more 'mav()' tests, including revising the PID gains. The net result is that I think the mav function does not initialize correctly, and thus always starts with a left-over error from the last use [not good if you want to drive two motors at the same rate to go straight or follow a repeatable turn. I suggest driving with 'motor()' function and using 'get_motor_position_counter()' to determine where to stop. rather than time.