This is machine translation

Translated by Microsoft
Mouseover text to see original. Click the button below to return to the English version of the page.

Note: This page has been translated by MathWorks. Click here to see
To view all translated materials including this page, select Country from the country navigator on the bottom of this page.

Probabilistic Roadmaps (PRM)

A probabilistic roadmap (PRM) is a network graph of possible paths in a given map based on free and occupied spaces. The robotics.PRM class randomly generates nodes and creates connections between these nodes based on the PRM algorithm parameters. Nodes are connected based on the obstacle locations specified in Map, and on the specified ConnectionDistance. You can customize the number of nodes, NumNodes, to fit the complexity of the map and the desire to find the most efficient path. The PRM algorithm uses the network of connected nodes to find an obstacle-free path from a start to an end location. To plan a path through an environment effectively, tune the NumNodes and ConnectionDistance properties.

When creating or updating the robotics.PRM class, the node locations are randomly generated, which can affect your final path between multiple iterations. This selection of nodes occurs when you specify Map initially, change the parameters, or update is called. To get consistent results with the same node placement, use rng to save the state of the random number generation. See Tune the Connection Distance for an example using rng.

Tune the Number of Nodes

Use the NumNodes property on the PRM object to tune the algorithm. NumNodes specifies the number of points, or nodes, placed on the map, which the algorithm uses to generate a roadmap. Using the ConnectionDistance property as a threshold for distance, the algorithm connects all points that do not have obstacles blocking the direct path between them.

Increasing the number of nodes can increase the efficiency of the path by giving more feasible paths. However, the increased complexity increases computation time. To get good coverage of the map, you might need a large number of nodes. Due to the random placement of nodes, some areas of the map may not have enough nodes to connect to the rest of the map. In this example, you create a large and small number of nodes in a roadmap.

Load a map file as a logical matrix, simpleMaps, and create an occupancy grid.

load exampleMaps.mat
map = robotics.OccupancyGrid(simpleMap,2);

Create a simple roadmap with 50 nodes.

prmSimple = robotics.PRM(map,50);
show(prmSimple)

Create a dense roadmap with 250 nodes.

prmComplex = robotics.PRM(map,250);
show(prmComplex)

The additional nodes increase the complexity but yield more options to improve the path. Given these two maps, you can calculate a path using the PRM algorithm and see the effects.

Calculate a simple path.

startLocation = [2 1];
endLocation = [12 10];
path = findpath(prmSimple,startLocation,endLocation);
show(prmSimple)

Calculate a complex path.

path = findpath(prmComplex, startLocation, endLocation);
show(prmComplex)

Increasing the nodes allows for a more direct path, but adds more computation time to finding a feasible path. Because of the random placement of points, the path is not always more direct or efficient. Using a small number of nodes can make paths worse than depicted and even restrict the ability to find a complete path.

Tune the Connection Distance

Use the ConnectionDistance property on the PRM object to tune the algorithm. ConnectionDistance is an upper threshold for points that are connected in the roadmap. Each node is connected to all nodes within this connection distance that do not have obstacles between them. By lowering the connection distance, you can limit the number of connections to reduce the computation time and simplify the map. However, a lowered distance limits the number of available paths from which to find a complete obstacle-free path. When working with simple maps, you can use a higher connection distance with a small number of nodes to increase efficiency. For complex maps with lots of obstacles, a higher number of nodes with a lowered connection distance increases the chance of finding a solution.

Load a map as a logical matrix, simpleMap, and create an occupancy grid.

load exampleMaps.mat
map = robotics.OccupancyGrid(simpleMap,2);

Create a roadmap with 100 nodes and calculate the path. The default ConnectionDistance is set to inf. Save the random number generation settings using the rng function. The saved settings enable you to reproduce the same points and see the effect of changing ConnectionDistance.

rngState = rng;
prm = robotics.PRM(map,100);
startLocation = [2 1];
endLocation = [12 10];
path = findpath(prm,startLocation,endLocation);
show(prm)

Reload the random number generation settings to have PRM use the same nodes. Lower ConnectionDistance to 2 m. Show the calculated path.

rng(rngState);
prm.ConnectionDistance = 2;
path = findpath(prm,startLocation,endLocation);
show(prm)

Create or Update PRM

Create or update your roadmap. To create the roadmap, call prm = robotics.PRM(map,__) or specify the Map property on the PRM object. Then, call the update, findpath, or show method. At this point, the nodes are randomly generated and the connections are made.

This roadmap changes only if you call update or change the properties in the PRM object. When properties change, any method (update, findpath, or show) called on the object triggers the roadmap points and connections to be recalculated. Because recalculating the map can be computationally intensive, you can reuse the same roadmap by calling findpath with different starting and ending locations.

Load the map, simpleMap, from a .mat file as a logical matrix and create an occupancy grid.

load('exampleMaps.mat')
map = robotics.BinaryOccupancyGrid(simpleMap,2);

Create a roadmap. Your nodes and connections might look different due to the random placement of nodes.

prm = robotics.PRM(map,100);
show(prm)

Call update or change a parameter to update the PRM nodes and connections.

update(prm)
show(prm)

The PRM algorithm recalculates the node placement and generates a new network of nodes.

References

[1] Kavraki, L.E., P. Svestka, J.-C. Latombe, and M.H. Overmars. "Probabilistic roadmaps for path planning in high-dimensional configuration spaces," IEEE Transactions on Robotics and Automation. Vol. 12, No. 4, Aug 1996 pp. 566—580.

See Also

| | |