| CPC B65G 47/905 (2013.01) [B25J 9/1697 (2013.01); G06T 7/62 (2017.01); G06T 7/70 (2017.01); H04N 13/239 (2018.05); H04N 13/296 (2018.05); G06T 2207/10028 (2013.01)] | 12 Claims |
|
1. A method for automatically placing a to-be-packed object into a container that defines an accommodation space, comprising steps of:
by a processing unit, obtaining an object dimension data piece that is related to dimensions of the to-be-packed object, and controlling a camera unit to capture an image of the accommodation space to obtain a space three-dimensional (3D) point cloud;
by the processing unit, obtaining an unoccupied area related to the accommodation space based on the space 3D point cloud;
by the processing unit, determining, based on the object dimension data piece and the unoccupied area, whether a first accommodation condition that corresponds to the to-be-packed object is satisfied; and
by the processing unit, upon determining that the first accommodation condition is satisfied, controlling a holding unit to place the to-be-packed object into the accommodation space at a position that corresponds to the unoccupied area,
wherein the step of obtaining the unoccupied area includes:
generating a first cross-section status data piece based on the space 3D point cloud, the first cross-section status data piece corresponding to a first cross section of the accommodation space that is parallel to a bottom surface of the container, and being indicative of an unoccupied area of the first cross section,
wherein the step of determining whether the first accommodation condition is satisfied includes:
determining whether the first cross section satisfies the first accommodation condition based on the object dimension data piece and the unoccupied area of the first cross section as indicated by the first cross-section status data piece,
wherein, upon determining that the first cross section satisfies the first accommodation condition, the processing unit controls the holding unit to place the to-be-packed object into the accommodation space at a position that corresponds to the unoccupied area of the first cross section,
wherein the step of obtaining an object dimension data piece includes:
controlling the camera unit to capture an image of the to-be-packed object, so as to obtain an object 3D point cloud; and
generating the object dimension data piece based on the object 3D point cloud,
wherein the image of the to-be-packed object is captured by the camera unit from below the to-be-packed object, and the image of the accommodation space is captured by the camera unit from above the container, and
wherein the camera unit includes a first depth camera disposed to face upwardly, and a second depth camera disposed to face downwardly, and the step of obtaining an object dimension data piece and controlling a camera unit to capture an image of the accommodation space includes:
controlling the first depth camera to capture the image of the to-be-packed object to obtain the object 3D point cloud; and
controlling the second depth camera to capture the image of the accommodation space to obtain the space 3D point cloud,
said method further comprising, before the step of obtaining an object dimension data piece, a step of:
by the processing unit, controlling the holding unit to move the to-be-packed object to a position above the first depth camera for the first depth camera to capture the image of the to-be-packed object.
|