.csv file read and computation of content of file

I have two .csv file which contains latitude and longitude in both the files. I need to read those values and calculate Haversine formula and plot.

[Manoj Pai]

lat1=15.850388;
lon1=74.498652;

lat2=15.850254;
lon2=74.498539;

lat3=15.850404;
lon3=74.498500;

R = 6371;                                   % Earth's radius in km
delta_lat = lat2 - lat1; % difference in latitude

%fprintf(' %d \n',delta_lat);
delta_lon = lon2 - lon1;        % difference in longitude

%fprintf(' %d \n',delta_lon);
a = sin(delta_lat/2)* sin(delta_lat/2) + cos(lat1) * cos(lat2) * sin(delta_lon/2)* sin(delta_lon/2);
c = 2 * atan2(sqrt(a), sqrt(1-a));
km = R * c;
fprintf(' %d \n',km);

delta_lat1 = lat3 – lat2;

delta_lon1 = lon3 – lon2;

a1 = sin(delta_lat1/2)* sin(delta_lat1/2) + cos(lat2) * cos(lat3) * sin(delta_lon1/2)* sin(delta_lon1/2);
c1 = 2 * atan2(sqrt(a1), sqrt(1-a1));
km1 = R * c1;
fprintf(' %d \n',km1);

Answer 1

[Walter Roberson]

latlong1 = csvread('FirstFile.csv');
latlong2 = csvread('SecondFile.csv');

lat1 = latlong1(:,1);
lon1 = latlong1(:,2);
lat2 = latlong2(:,1);
lon2 = latlong2(:,2);

R = 6371;                                   % Earth's radius in km
delta_lat = lat2 - lat1; % difference in latitude

%fprintf(' %d \n',delta_lat);
delta_lon = lon2 - lon1;        % difference in longitude

%fprintf(' %d \n',delta_lon);
a = sin(delta_lat/2) .* sin(delta_lat/2) + cos(lat1) .* cos(lat2) .* sin(delta_lon/2) .* sin(delta_lon/2);                      %notice each * was replaced by .*
c = 2 * atan2(sqrt(a), sqrt(1-a));
km = R * c;
fprintf(' %d \n',km);

[Manoj Pai]

Thank you Walter

[Manoj Pai]

I need to plot these values. Please guide me..

[Manoj Pai]

f1=figure(1);
cla;
hold on;
 plot(lon1,lat1,lon2,lat2,'k-','marker','+')
 %plot(lon2,lat2,'.g','MarkerSize',10)
 plot_google_map
%is this way we can do?

[Walter Roberson]

That looks possible. I do not know plot_google_map though.

[Manoj Pai]

ok. It works plot_google_map. If we run the plot_google_map function. The .csv file has (ie FirstFile.csv and SecondFile.csv) contains multiple data(it contains of about 500 values of lat and long). I need to calculate Haversine formula for each data(lat long) of the .csv file. Does the solution provided above by you works?

[Walter Roberson]

Yes. With the change from * to .* that I pointed out, all of the values will be computed at the same time, provided that you are comparing the corresponding entries in each set. (If you want to take the distance from each member of the first file to every member of the second file, and so on, then the code would have to change a bit.)

[Manoj Pai]

Samples of csv file data First file lat lon 15 74 16 75 .. ..

Second file lat lon 17 76 18 77 .. .. I need to calculate first entry of lat lon of firstfile.csv (15 74) with first entry of secondfile.csv(17,76) and second entry of firstfle.csv (16 75) with second entry of secondfile.csv(18 77) and so on... Could you please tell me what changes needs to be done?

[Manoj Pai]

These are the two .csv files

[Walter Roberson]

"and second entry of firstfle.csv (16 75) with second entry of secondfile.csv(18 77)"

That is already what the code does.

[Manoj Pai]

Thank you...

[Manoj Pai]

beep; msgbox('Danger, Will Robinson!') I need this message when there is the intersection of lat, long of both the files. I have shared the output of the code. The point at which there is intersection an alert message should be popped up.

[Walter Roberson]

if any(km == 0); beep; msgbox('Danger, Will Robinson!'); end

[Manoj Pai]

I have those lat, long values in radians.

How to calculate the trigonometric functions for radians values.

delta_lat = lat2 - lat1; % difference in latitude
%fprintf(' %d \n',delta_lat);
delta_lon = lon2 - lon1;        % difference in longitude

a = sin(delta_lat/2) .* sin(delta_lat/2) + cos(lat1) .* cos(lat2) .* sin(delta_lon/2) .* sin(delta_lon/2);                      
c = 2 * atan2(sqrt(a), sqrt(1-a));

[Walter Roberson]

sin() and cos() and atan2() automatically work with radians.

You would have needed to do something special to work with degrees -- you would have had to use sind() and cosd() and atan2d()

[Manoj Pai]

ok Thank You

[Manoj Pai]

How to plot lat1 lon1 and lat2 lon2 on a single graph

[Walter Roberson]

 plot(lon1,lat1,lon2,lat2,'k-','marker','+')
 hold on
 plot(lon2,lat2,'.g','MarkerSize',10)

[Manoj Pai]

Thanks a lot

[Manoj Pai]

In this file first row consists of title as lat1, lon1...... Should I need to do some changes in this part of the code?

lat1 = latlong1(:,1);
lon1 = latlong1(:,2);
lat2 = latlong2(:,1);
lon2 = latlong2(:,2);

[Walter Roberson]

latlong1 = csvread('FirstFile.csv', 1, 0);

Note that this requires R2014a or so (not sure of exact first release)

[Manoj Pai]

latlong1= csvread('C:\Users\Desktop\FirstFile .csv',1,0);

[Manoj Pai]

Can we use this for file in different directory

[Walter Roberson]

Sure, reading from a different directory is no problem, in exactly the way you posted. Just watch out: I see a space between 'FirstFile' and '.csv' that should probably not be there.

[Manoj Pai]

I am new to Matlab.Thank you I learnt a lot from you.

[Manoj Pai]

clear
clc

HV = csvread('FirstFile.csv',1,0);
RV = csvread('SecondFile.csv',1,0);
lat1 = HV(:,1);
lon1 = HV(:,2);
speed1 = HV(:,3);
heading1 = HV(:,4);

lat2 = RV(:,1);
lon2 = RV(:,2);
speed2 = RV(:,3);
heading2 = RV(:,4);

R = 6378.137;                                   %Earth's radius in km
KMPH_TO_MPS = 0.277778;                %conversion from kmph to mps
iteration = 200;                                %iteration for 200 values 
T = 1;
for i=0:iteration

% %Calculation of future points for HV
   Vx = sind(heading1).*speed1;
   Vy = cosd(heading1).*speed1 ;
   Lat2_HV = (lat1 + (T * Vy/111));% T is in hours
   lat1=Lat2_HV;

      Long2_HV = lon1 +( T * (Vx / cosd(Lat2_HV))/111);% 1 degree of lattitude/longitude corresponds to 111km
      lon1=Long2_HV;

%Calculation of fiture points for RV
    Vx1 = sind(heading2).* speed2;
    Vy1 = cosd(heading2).*speed2;
    Lat2_RV = (lat2 + (T * Vy1/111)); % T is in hours
    lat2=Lat2_RV;

      Long2_RV = lon2 +( T * (Vx1 / cosd(Lat2_RV))/111); % 1 degree of lattitude/longitude corresponds to 111km
      lon2=Long2_RV;

      delta_lat  = (Lat2_HV  -  Lat2_RV ) ;          % Difference between latitudes of HV and RV

  delta_lon = (Lon2_HV  -  lon2_RV) ;       % Difference between longitudes of HV and RV
  
  a = sind(delta_lat/2) .* sind(delta_lat/2) + cosd(Lat2_RV) .* cosd(Lat2_HV) .* sind(delta_lon/2) .* sind(delta_lon/2);     %notice each * was replaced by .*
  c = 2 * atan2(sqrt(a), sqrt(1-a));
  km = R * c;
  m = km * 1000;
     plot(lat1,lon1,'*',lat2,lon2,'+',Lat2_HV,Lat2_RV,'b--',Lon2_HV,Lon2_RV,'r--'); hold on

      drawnow
    fprintf('Distance between vehicles= %f \n',m);

      if(m<=30)  % Checking safety distance for alert condition

% Calculation for relative velocity
     HV1=((speed1)*(cosd(heading1)));
  
   HV2=((speed1)*(sind(heading1 )));

   RV1=((speed2)*(cosd(heading2)));
  
   RV2=((speed2)*(sind(heading2)));
   
   s1=(HV1)+(RV1);

   s2=(HV2)+(RV2);

   R_speed=(s1.^2)+(s2.^2);
   
   R_speed1=(R_speed.^(1/2));

   fprintf('RELATIVE_SPEED= %f \n',R_speed1); %Display relative velocity of HV and RV

        %   calculation of TTC with predicted trajectory of HV and RV with relative speed of HV and RV

    TTC=dm2/(R_speed1* KMPH_TO_MPS);
    fprintf('TTC= %f \n',TTC); 
    msgbox('IMA ALERT');
      fprintf('ALERT');

      else

  fprintf('safe');

      end
  end

%plot(lon1,lat1,lon2,lat2,'k-','marker','+')
%hold on
%plot(lon2,lat2,'.g','MarkerSize',10)

[Manoj Pai]

Error using  + 
Matrix dimensions must agree.

Error in calculation (line 29)
    Long2_HV = lon1 +( T * (Vx / cosd(Lat2_HV))/111);% 1 degree of lattitude/longitude corresponds to 111km

I am getting this error

[Walter Roberson]

The files that you posted first time around only have two columns. You posted a new version of FirstFile.csv with more columns, but you did not post a new version of SecondFile.csv

[Manoj Pai]

Yes it contains 4 columns in both the files.. lat1, lon1, speed1, heading1

[Walter Roberson]

You had / in two places where you needed ./

You had * in some places where you needed .*

However: you are treating m (the vector of distances) as if it were a scalar, but it is a vector. When you go to print out the distances, you are printing out 599 distances at the same time, which is pretty much useless. When you test m<=30 you are testing whether all m<=30 because of the way that vectorized "if" works; you can change that to any(m<=30) to at least be more correct.

[Manoj Pai]

In the above code I am predicting the future points from previous points(which is in csv file) and if future points distance is less than 30m then printing alert and msgbox. Vx = sind(heading1).*speed1; Vy = cosd(heading1).*speed1 ; Lat2_HV = (lat1 + (T .* Vy/111));% T is in hours lat1=Lat2_HV;

      Lon2_HV = lon1 +( T .* (Vx ./ cosd(Lat2_HV))./111);% 1 degree of lattitude/longitude corresponds to 111km
      lon1=Lon2_HV;

%Calculation of fiture points for RV
    Vx1 = sind(heading2).* speed2;
    Vy1 = cosd(heading2).*speed2;
    Lat2_RV = (lat2 + (T .* Vy1./111)); % T is in hours
    lat2=Lat2_RV;

      Lon2_RV = lon2 +( T * (Vx1 ./ cosd(Lat2_RV))./111); % 1 degree of lattitude/longitude corresponds to 111km
      lon2=Lon2_RV;