LiDAR Sim

🎆 A 2D LiDAR visualization for simulated robots in MATLAB at the Georgia Tech Robotarium

〰️ Ability to modify distance, angular resolution, and Gaussian distortion of points

🗻 Works with any N-vertice shape and multiple robots simultaneously

Usage:

LIDAR(x, N, ang_Res, dist, obs, error)

• x is the current robot's x,y position
• N is the number of active robots
• ang_Res is how many 360/N rays to be generated
• dist is the maximum distance range of a ray
• obs are predefined obstacles given by its vertices
• error is defined as either 'None' or 'Gaussian'

returns

• output: a matrix of all calculated intersection points [x,y]

Refer to the go_to_pose.m file for an example of the function call, obstacle initialization, and plotting routine

Plotting:

• Initialize Plot of LIDAR points before experiment loop:

RGB = [255 0 0]/256;  
scan = [0,0];  
h = scatter(scan(:,1), scan(:,2), 72, RGB, 'filled'); 

• In loop plotting update:

[numP,numCols] = size(scan);     
for g = 1:numP
    set(h,'XData',scan(:,1),'YData',scan(:,2));      
end

Warning:

• The real-time run speed of this function is variable with the amount of intersections that are computed and plotted, thus simpler (less robots, less obstacles) experiments are more likely to run
• As N number of robots increase, ang_res and dist need to be lowered to ensure real-time operation on Robotarium
• See the following table for sample runs and respective real-time run speed

Point Cloud Test:

N	ang_Res	dist	Time w/ Plotting (s)	Time w/o Plotting (s)
1	1	0.5	0.0295	0.0331
2	1	0.5	0.0414	0.0347
3	1	0.5	0.1094	0.0829
1	5	0.5	0.0138	0.0326
2	5	0.5	0.0291	0.0278
3	5	0.5	0.0350	0.0233
5	5	0.5	0.0428	0.0450
6	18	0.5	0.0332	0.0331
8	18	0.5	0.0513	0.0452
10	36	0.5	0.0402	0.0353