Unmanned aerial vehicle (UAV) and method for operating the UAV
An improved UAV system and methods for operation in an inventory management system. The methods include generating a three dimensional (3D) map and estimating a position and orientation of the UAV based upon this map; autonomously navigating the UAV in the environment by using the generated 3d map in conjunction with the position and the orientation of the UAV; performing static and dynamic obstacle avoidance in the environment using collision avoidance; and finding the optimal path from a source node to a destination node within the environment.
1. An unmanned aerial vehicle (UAV) device comprising:
a processor ( 108 ) communicatively coupled with at least one camera ( 102 ), a memory ( 106 ), at least one sensor ( 112 ), an inertial measurement unit ( 118 ), and a flight controller ( 120 ), the processor ( 108 ) adapted to receive and execute stored program instructions to perform the method steps comprising:
generating a three-dimensional (3D) map of an operating environment;
estimating a spatial position of the UAV device as a location current node and an orientation of the UAV device within the environment;
finding an optimal path of travel of the UAV device from a first node within the environment to a second node within the environment by automatically computing waypoints therebetween as the UAV device travels;
navigating the UAV device within the environment along the optimal path from the first node to the second node, autonomously and dynamically, using a path planning estimate in combination with the waypoints, spatial position estimate, and orientation estimate, while avoiding obstacles encountered along the path within the environment;
utilizing an obstacle detection algorithm including a minimum allowed distance value as a minimum allowed distance from an obstacle; and
automatically slowing the UAV device as it approaches an obstacle in response to this minimum allowed distance value.
2. The device of claim 1 , the method steps further comprising: accepting, from the inertial measurement unit ( 118 ) an estimate of the initial UAV device position to improve the accuracy of an odometry estimate.
3. The device of claim 1 , the method steps further comprising: using a graph optimizer algorithm ( 110 ) receiving input from a stereo camera ( 102 ) for optimization of the spatial position estimate.
4. The device of claim 1 , the method steps further comprising: utilizing a loop-closure detection algorithm ( 116 ) to allow the UAV device to return to a previous spatial position.
5. The device of claim 1 , the method steps further comprising: receiving input data from a stereo camera ( 102 ) and generating a UAV odometry data and point cloud data from same; utilizing the point cloud data in generation of the 3D map.
6. The device of claim 1 , the method steps further comprising: receiving a sequence of image frame data; and managing the memory ( 106 ) utilization by buffering and maintaining a most recent image frame data in a current memory buffer.
7. The device of claim 1 , the method steps further comprising: receiving an image frame data; and building a point cloud around the image frame data edges to indicate the device location within the image frame.
8. The device of claim 1 , the method steps further comprising: determining the waypoints along the optimal path, wherein the distance between waypoints is reduced in proximity to an obstacle or a tight turn radius on the optimal path; and automatically varying the speed of travel of the UAV device based upon the spacing of the waypoints.
9. The device of claim 1 , the method steps further comprising: creating a point cloud of the operating environment; creating the 3D grid data from the point cloud; and determining the optimal path using an A* Algorithm operating on a two-dimensional grid data.
10. A method of operating an unmanned aerial vehicle (UAV), the method steps comprising:
generating, with a UAV device ( 100 ) processor ( 108 ) in communication with a memory ( 106 ), a three-dimensional (3D) grid map of an operating environment using data from at least one camera ( 102 );
estimating, with data from an inertial measurement unit ( 118 ), a spatial position of the UAV device as a location current node and an orientation of the UAV device within the environment;
finding an optimal path of travel of the UAV device from a first node within the environment to a second node within the environment by the processor automatically computing waypoints therebetween as the UAV device travels;
navigating, with a flight controller ( 120 ), the UAV device within the environment along the optimal path from the first node to the second node, autonomously and dynamically, using a path planning estimate in combination with the waypoints, spatial position estimate, and orientation estimate, while avoiding obstacles encountered along the path within the environment;
utilizing an obstacle detection algorithm including a minimum allowed distance value as a minimum allowed distance from an obstacle; and
automatically slowing the UAV device as it approaches an obstacle in response to this minimum allowed distance value.
11. The method of claim 10 , the method steps further comprising: accepting, from the inertial measurement unit ( 118 ) an estimate of the initial UAV device position to improve the accuracy of an odometry estimate.
12. The method of claim 10 , the method steps further comprising: using a graph optimizer algorithm ( 110 ) receiving input from a stereo camera ( 102 ) for optimization of the spacial position estimate.
13. The method of claim 10 , the method steps further comprising: utilizing a loop-closure detection algorithm ( 116 ) to allow the UAV device to return to a previous spatial position.
14. The method of claim 10 , the method steps further comprising: receiving input data from a stereo camera ( 102 ) and generating a UAV odometry data and point cloud data from same; and utilizing the point cloud data in generation of the 3D map.
15. The method of claim 10 , the method steps further comprising: receiving a sequence of image frame data; and managing the memory ( 106 ) utilization by buffering and maintaining a most recent image frame data in a current memory buffer.
16. The method of claim 10 , the method steps further comprising: receiving an image frame data; and building a point cloud around the image frame data edges to indicate the UAV device ( 100 ) location within the image frame.
17. The method of claim 10 , the method steps further comprising: determining the waypoints along the optimal path, wherein the distance between waypoints is reduced in proximity to an obstacle or a tight turn radius on the optimal path; and automatically varying the speed of travel of the UAV device based upon the spacing of the waypoints.
18. The method of claim 10 , the method steps further comprising: creating a point cloud of the operating environment; creating the 3D grid data from the point cloud; and determining the optimal path using an A* Algorithm operating on a two-dimensional grid data.