|Build occupancy map from lidar scans|
|Check locations for free, occupied, or unknown values|
|Get occupancy value of locations|
|Inflate each occupied grid location|
|Insert ray from laser scan observation|
|Insert 3-D points or point cloud observation into map|
|Move map in world frame|
|Convert occupancy grid to double matrix|
|Compute cell indices along a ray|
|Find intersection points of rays and occupied map cells|
|Set occupancy value of locations|
|Sync map with overlapping map|
|Show grid values in a figure|
|Integrate probability observations at locations|
Occupancy Maps offer a simple yet robust way of representing an environment for robotic applications by mapping the continuous world-space to a discrete data structure. Individual grid cells can contain binary or probabilistic information, where 0 indicates free-space, and 1 indicates occupied space. You can build up this information over time using sensor measurements and efficiently store them in the map. This information is also useful for more advanced workflows, such as collision detection and path planning.
This example shows how to create an egocentric occupancy map from the Driving Scenario Designer app. This example uses obstacle information from the vision detection generator to update the egocentric occupancy map.
buildMap function takes in lidar scan readings and associated poses to build an occupancy grid as
lidarScan objects and associated
[x y theta] poses to build an
This example shows how to reduce the drift in the estimated trajectory (location and orientation) of a monocular camera using 3-D pose graph optimization. In this example, you build an occupancy map from the depth images, which can be used for path planning while navigating in that environment.