Main Content

controllerVFH3D

Avoid obstacles using 3D vector field histogram

Since R2022b

Description

The controllerVFH3D System object™ enables a UAV to avoid obstacles, based on sensor data, by using 3D vector field histograms (3DVFH). The object computes an obstacle-free direction using the sensor-data-based positions of obstacles, the UAV position, the UAV orientation, and the target position.

controllerVFH3D uses the 3DVFH+ algorithm[1] to compute an obstacle-free direction.

To find an obstacle-free direction:

  1. Create the controllerVFH3D object and set its properties.

  2. Call the object with arguments, as if it were a function.

To learn more about how System objects work, see What Are System Objects?

Creation

Description

example

vfh3D = controllerVFH3D returns a 3-D vector field histogram object, vfh3D, that computes a desired direction, a desired yaw, and a status output using the 3DVFH+ algorithm.

vfh3D = controllerVFH3D(Name=Value) returns a vector field histogram object with properties specified by one or more name-value arguments. Properties not specified retain their default values.

Properties

expand all

Unless otherwise indicated, properties are nontunable, which means you cannot change their values after calling the object. Objects lock when you call them, and the release function unlocks them.

If a property is tunable, you can change its value at any time.

For more information on changing property values, see System Design in MATLAB Using System Objects.

Histogram grid resolution, specified as 1, 3, 5, 6, 10, 15, 18, 30, 45, or 60. All values are in degrees.

Example: HistogramResolution=10

Data Types: single | double

Histogram window size, specified as a positive odd integer. The histogram window size determines the angular width of an obstacle-free opening in the azimuth and elevation directions. This value is unitless.

Example: WindowSize=3

Data Types: single | double

Threshold for computing the histogram, specified as a positive integer. This value specifies the minimum number of obstacle points that must be in a histogram cell for the cell to be considered as an obstacle. If a cell contains fewer than this number of obstacle points, the cell is considered as obstacle-free.

Example: HistogramThreshold=2

Data Types: single | double

Maximum age of remembered obstacle points, specified as a numeric scalar. This value specifies the number of time steps prior to the current step from which the object remembers the obstacle points.

Example: MaxAge=1

Data Types: single | double

Limits of the range sensor, specified as a vector of the form [min max], with values in meters.

Example: DistanceLimits=[0.3 15]

Data Types: single | double

Horizontal field of view limits of the range sensor, specified as a vector of the form [min max], with values in degrees.

Example: HorizontalSensorFOV=[-50 50]

Data Types: single | double

Vertical field of view limits of the range sensor, specified as a vector of the form [min max], with values in degrees.

Example: VerticalSensorFOV=[-20 20]

Data Types: single | double

Sensor mounting location on the UAV, specified as a vector of the form [x y z], with values in meters.

Example: SensorLocation=[0 0 0]

Data Types: single | double

Orientation of the sensor mounted on the UAV, specified as a vector of the form [roll pitch yaw], with values in degrees.

Example: SensorOrientation=[0 0 0]

Data Types: single | double

Radius of the UAV, specified as a numeric scalar in meters. This value specifies the radius of the smallest circle that can circumscribe the UAV. The object uses the vehicle radius to account for vehicle size when computing the obstacle-free direction.

Example: VehicleRadius=0.5

Data Types: single | double

Safety distance between UAV and obstacle, specified as a numeric scalar in meters. This value specifies the space to account for between the UAV and obstacles in addition to the vehicle radius. The object uses both the vehicle radius and safety distance to compute the obstacle-free direction.

Example: SafetyDistance=2

Data Types: single | double

Cost function weight for moving in the target direction, specified as a numeric scalar. To prioritize following a target direction, set this weight to a value greater than the sum of CurrentDirectionWeight and PreviousDirectionWeight. To remove the target direction from consideration in the cost function, set this weight to 0.

Example: TargetDirectionWeight=0

Data Types: single | double

Cost function weight for moving in the current direction, specified as a numeric scalar. Higher values of this weight can produce more efficient paths. To remove the current direction from consideration in the cost function, set this weight to 0.

Example: CurrentDirectionWeight=0

Data Types: single | double

Cost function weight for moving in the previous direction, specified as a numeric scalar. Higher values of this weight can produce smoother paths. To remove the previous direction from consideration in the cost function, set this weight to 0.

Example: PreviousDirectionWeight=1

Data Types: single | double

Usage

Description

example

[desiredDirection,desiredYaw,status] = vhf3D(position,orientation,obstaclePoints,targetPosition) finds an obstacle-free direction and yaw, using the 3DVFH+ algorithm, for the input UAV position, UAV orientation, sensor-data-based positions of obstacles, and target position. This syntax also returns the status status of the obstacle-free direction.

Input Arguments

expand all

Position of the UAV, specified as a vector of the form [x; y; z], in meters.

Example: [1; 1; 1]

Data Types: single | double

Orientation of the UAV, specified as a quaternion vector of the form [w; x; y; z].

Example: [1; 0; 0; 0]

Data Types: single | double

Positions of the obstacles, specified as an N-by-3 matrix with rows of the form [x y z], in meters. N is the number of obstacle points.

Example: [1 1 1; 2 2 2]

Data Types: single | double

Target position, specified as a vector of the form [x; y; z], in meters.

Example: [2; 3; 4]

Data Types: single | double

Output Arguments

expand all

Desired direction, returned as a vector of the form [x; y; z], in meters.

Data Types: single | double

Desired yaw, returned as numeric scalar in the range [-pi, pi], in radians.

Data Types: single | double

Status of the obstacle-free direction, returned as 0, 1, 2, or 3.

  • 0 — The object finds an obstacle-free direction.

  • 1 — The object does not find an obstacle-free direction.

  • 2 — The object finds an obstacle-free direction, but the direction is close to an obstacle.

  • 3 — The object does not find an obstacle-free direction, and the direction is close to an obstacle.

Data Types: uint8

Object Functions

To use an object function, specify the System object as the first input argument. For example, to release system resources of a System object named obj, use this syntax:

release(obj)

expand all

showDisplay 3D vector field histogram
stepRun System object algorithm
releaseRelease resources and allow changes to System object property values and input characteristics
resetReset internal states of System object

Examples

collapse all

Create a controllerVFH3D object.

vfh3D = controllerVFH3D;

Create obstacles.

az = [-60:-20 20:60]*(pi/180);
el = (-30:30)*(pi/180);
[El,Az] = meshgrid(el,az);

Specify the distances of the obstacles from the sensor, and convert to Cartesian coordinates.

obstacleDist = linspace(15,20,numel(El(:)));
[xSensor,ySensor,zSensor] = sph2cart(Az(:),El(:),obstacleDist');

Align the sensor and histogram frames.

vfh3D.SensorOrientation = [-180 0 0];

Specify the sensor range limits.

vfh3D.DistanceLimits = [0.2 25];

Specify the current UAV position and orientation, the locations of obstacles, and the target position for the UAV.

uavPosition = [0; 0; 0];
uavOrientation = [1; 0; 0; 0];
sensorPoints = [xSensor ySensor zSensor];
targetPosition = [20; 0; 0];

Compute an obstacle-free direction and desired yaw for the UAV, and return the status of the obstacle-free direction.

[desiredDirection,desiredYaw,status] = vfh3D(uavPosition, ...
                                             uavOrientation, ...
                                             sensorPoints, ...
                                             targetPosition);

Visualize the default histogram of the calculated direction.

show(vfh3D)
axis equal

Visualize the 2D memory histogram, 2D inflated histogram, and cost matrix.

figure
ax(1) = subplot(3,1,1);
ax(2) = subplot(3,1,2);
ax(3) = subplot(3,1,3);
show(vfh3D, ...
     Parent=ax, ...
     PlotsToShow=["2D Memory Histogram","2D Inflated Histogram","Cost Matrix"])
axis(ax,"equal")
xlim(ax,"tight")

References

[1] Baumann, Tanja. “Obstacle Avoidance for Drones Using a 3DVFH* Algorithm.” Master's thesis, Eidgenössische Technische Hochschule Zürich, 2018.

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.

Version History

Introduced in R2022b

See Also

Objects

Functions