Main Content


Estimate cost of propagating to target state

Since R2021b



    h = distance(mobileProp,q1,q2) estimates the cost of propagating from one state to another. The DistanceEstimator property of the state propagator defines the distance metric for approximating cost.


    collapse all

    Create a state propagator and specify the distance metric for estimating propagation cost.

    propagator = mobileRobotPropagator(DistanceEstimator="dubins");

    Create a Dubins state space.

    dubinsSpace = stateSpaceDubins([0 25; 0 25; -pi pi]);

    Update the state space of the state propagator using the created state space.

    propagator.StateSpace = dubinsSpace;

    Create a navPath object based on multiple waypoints in a Dubins space.

    path = navPath(dubinsSpace);
    waypoints = [8 10 pi/2;
                 7 14 pi/4;
                 10 17 pi/2;
                 10 10 -pi];

    Interpolate that path so that it contains exactly 250 points.

    numStates = 250;

    Extract the sequence of motions from the path.

    q1 = path.States(1:end-1,:); % Initial states
    q2 = path.States(2:end,:);   % Final states

    Estimate the cost of propagating to target state.

    cost = distance(propagator,q1,q2);

    Generate a series of control commands and number of steps to move from the current state q1 with control command u toward the target state q2.

    u = zeros(size(q1,1),propagator.NumControlOutput);
    steps = zeros(size(q1,1),1);
    for i = 1:size(q1,1)
        [u(i+1,:),steps(i)] = sampleControl(propagator,q1(i,:),u(i,:),q2(i,:));

    Create a control-based path object with the specified state propagator and a sequence of specified states, controls, targets, and durations.

    states = path.States;
    controls = u(2:end,:);
    targets = q2;
    durations = steps*propagator.ControlStepSize;
    path2 = navPathControl(propagator,states,controls,targets,durations);

    Visualize the results.

    grid on
    axis equal
    hold on

    Input Arguments

    collapse all

    Mobile robot state propagator, specified as a mobileRobotPropagator object.

    Initial states, specified as an n-by-s matrix. n is the number of states and s is the size of the state vector.

    Final states, specified as an n-by-s matrix. n is the number of states and s is the size of the state vector.

    Output Arguments

    collapse all

    Cost values, returned as an n-element vector. n is the number of q1 and q2 pairs.

    You can use the cost values returned by this function to find the nearest neighbor for sampled target states when planning a path.

    Version History

    Introduced in R2021b