System and method for load-carrier pose estimation
A system for Load-carrier Pose estimation. The system includes a Load-carrier for supporting a load with Load-carrier handling indicators or Key-points, and a driving vehicle. The Pose represents a relative position and orientation of the Load-carrier with reference to the vehicle. The system includes a 2D sensor and a 3D sensor. The system is arranged to detect 2D Keypoints in images of the Load-carrier obtained with the 2D sensor, and to match these 2D Keypoints with 3D data captured with the 3D sensor to establish a 3D position and orientation of the detected Keypoints.
1. A system for Load-carrier Pose estimation, the system comprising:
a driving vehicle;
a Load-carrier for supporting a load, wherein the Load-carrier comprises Load-carrier handling indicators or Keypoints; and
a 2D sensor and a 3D sensor, wherein the system is arranged to carry out a first estimation phase in which a Load-carrier detection is carried out, followed by a second estimation phase wherein a Pose estimation is applied, the Pose representing a relative position and orientation of the Load-carrier with reference to the vehicle, and wherein the second estimation phase is embodied with 2D key point detection and 3D matching of such 2D Keypoints, and wherein in a third estimation phase the system employs Pose estimation using 3D template matching initialized by the Pose estimation resulting from the second phase.
2. The system according to claim 1 , wherein the system is arranged to detect 2D Keypoints in images of the Load-carrier obtained with the 2D sensor, and to match these 2D Keypoints with 3D data captured with the 3D sensor to establish a 3D position and orientation of the detected Keypoints.
3. The system according to claim 1 , wherein the 3D position and orientation of the detected Keypoints of the Load-carrier in the place for a final alignment is supplied to a processor of the system that carries out the 3D template matching optimization.
4. The system according to claim 1 , wherein the system comprises a 3D point cloud template of the Load-carrier, wherein the system is arranged to superimpose the 3D point cloud template onto 3D data derived with the 3D sensor, which 3D data represents the Load-carrier in 3D space, and wherein said superposition enables computing the Load-carrier's position.
5. The system according to claim 1 , wherein the 2D sensor and/or the 3D sensor are mounted on the vehicle, thus providing said sensors with a fixed 6 degrees of freedom offset for use in a 2D to 3D coordinate matrix transformation.
6. The system according to claim 1 , further comprising an IMU sensor.
7. The system according to claim 1 , wherein the system operates in real-time enabling to estimate the Pose at least once while the vehicle is moving.
8. The system according to claim 1 , further comprising an image processor connected to the 2D sensor for detecting the Load-carrier and/or 2D Keypoints which generates image data at a frequency of more than 30 Hz.
9. The system according to claim 8 , wherein the image processor combines the detection of the 2D Key-points with Load-carrier detection.
10. The system according to claim 8 , wherein the image processor detects the Load-carrier and the 2D Keypoints using machine learning/deep learning by applying the same deep neural network for detecting the Load-carrier and the 2D Keypoints simultaneously.
11. The system according to claim 1 , wherein the Load-carrier is a pallet.
12. The system according to claim 1 , wherein the vehicle is a pallet lifter or forklift.
13. A method for Load-carrier Pose estimation of a Load-carrier, wherein the Load-carrier comprises Load-carrier handling indicators or Keypoints, wherein the Pose represents a relative position and orientation of the Load-carrier with reference to a vehicle, the method comprising the steps of:
employing a 2D sensor and a 3D sensor;
executing a first estimation phase in which a Load-carrier detection is carried out;
executing a second estimation phase subsequent to the first estimation phase applying a Pose estimation and employing 2D key point detection and 3D matching of such 2D Keypoints;
providing an image processor connected to the 2D sensor for detecting the Load-carrier and/or the 2D Keypoints which generates image data at a frequency of more than 30 Hz;
arranging that the image processor combines the detection of the 2D Keypoints with Load-carrier detection; and
detecting the 2D Keypoints and the Load-carrier by using the same deep neural network simultaneously.
14. The method according to claim 13 further comprising the steps of:
detecting 2D Key-points in images of the Load-carrier obtained with the 2D sensor; and
matching these 2D Keypoints with 3D data captured with the 3D sensor to establish a 3D position and orientation of the detected Keypoints.
15. The method of claim 13 further comprising the step of executing a third estimation phase employing Pose estimation using 3D template matching initialized by the Pose estimation resulting from the second estimation phase.
16. The method of claim 15 further comprising the step of: providing the 3D position and orientation of the detected Keypoints of the Load-carrier in the place for a final alignment to initialize the 3D template matching optimization.
17. The method of claim 15 further comprising the step of superimposing a template of the Load-carrier onto 3D data derived with the 3D sensor, which 3D data represents the Load-carrier in 3D space, wherein said superposition enables computing the Load-carrier's position.
18. The method according to claim 13 further comprising the step of estimating the Pose while the vehicle is moving.
19. The method according to claim 13 , further comprising the step of using the image processor to detect the Load-carrier and/or 2D Keypoints using machine learning/deep learning.