Insert Date Column Based off Known Start and Stop Time

Hello,

I have processed data that covers a 5 month period and have 3 columns of data that give me the x, y, and z rotation of an object. The data is divided by data points and not by time. I know the start and stop time for the data logging. What is the best way to create a date field that I can fill in the start and end time? Ideally I'd like to plot these values with known date values instead of just the data point value on the x-axis.

[the cyclist]

It would be a lot easier to understand your question if you posted a small (but representative) example of the input/output you expect.

Don't just describe it. Post actual code, or upload a MAT file.

Answer 1

[Peter Perkins]

It sounds like maybe you want a table that contains two datetime variables as well as x, y, and z, but as the cyclist says, you're gonna have to provide more details.

Answer 2

[Daniel Devereaux]

Hi. I've attached a MAT file that covers 3 hours of data. The unit collects data points at 100hz for 10min every 3 hrs. So it runs for 10 mins after the following start time with each data point representing .01 of a second.

Start time: 6:08:00 PM UTC 2016-05-22

How would I be able to plot these three columns of data (x,y,z) against time referenced to this start time in that increment?

[Peter Perkins]

Here's some code that works in R2016b:

>> t0 = datetime('6:08:00 PM UTC 2016-05-22','InputFormat','hh:mm:ss a ''UTC'' yyyy-MM-dd');
>> t0.Format = 'yyyy-MM-dd HH:mm:ss.SSS';
>> d = (minutes(0):milliseconds(10):minutes(10))';
>> t = t0 + d;
>> xyz = cumsum(randn(length(t),3));
>> tt = array2timetable(xyz,'RowTimes',t,'VariableNames',{'x' 'y' 'z'})
tt =
              Time               x          y           z    
    _______________________    ______    ________    ________
    2016-05-22 18:08:00.000    0.2992    -0.95717     0.01993
    2016-05-22 18:08:00.010    1.7475      -1.054    -0.52164
    2016-05-22 18:08:00.020    2.2103     -0.2601    -0.66848
    2016-05-22 18:08:00.030    2.7189     0.69421    -0.41496
    2016-05-22 18:08:00.040     2.381     0.28662      -1.789
    2016-05-22 18:08:00.050    4.5419     0.76694    -0.78308
    2016-05-22 18:08:00.060     3.748      1.9123     -2.2305
    2016-05-22 18:08:00.070    2.7249      0.9289     -3.7441
    2016-05-22 18:08:00.080    2.4665       2.068     -3.9346
    2016-05-22 18:08:00.090    1.3154      1.2933     -4.4244
[snip]

Plot all three vars on one axis:

plot(tt.Time,tt.Variables)
legend(tt.Properties.VariableNames)

Plot on separate axes (using ttplot from the File Exchange):

ttplot(tt)

Both can also be done prior to R2016b, using a table, not a timetable.

Hard to know how you want to handle those 2hr 50min gaps, the above only plots one segment.