Terrain verification system and method for pallet drop
A work machine includes a terrain verification system for dropping a pallet on a land surface at a destination. Operating autonomously or semi-autonomously, the work machine includes a range sensor, an image sensor, and an inertial sensor to assess the slope and roughness of the land surface on approach to the destination. The terrain verification system generates a ground map of the destination from range data received by the range sensor and generates a roughness metric indicative of the surface. Image data received by the image sensor is used to verify the roughness metric and, when obstructions are found from the image data, the system moves the work machine over the land surface and evaluates pitch and roll data from the inertial sensor to determine the condition of the destination for dropping the pallet.
1. A movable work machine, comprising:
a chassis;
a range sensor, coupled to the chassis, configured to generate range data indicative of topology of a land surface at a destination of the movable work machine;
an image sensor, coupled to the chassis, configured to generate image data indicative of the topology;
an inertial sensor, coupled to the chassis, configured to generate orientation data for the movable work machine during traversal of the land surface; and
a terrain verification system, comprising:
a terrain mapper configured to convert the range data into a ground map of the land surface;
an obstruction detector configured to identify one or more obstructions on the land surface from the image data;
a localization system configured to evaluate orientation of the movable work machine from at least the orientation data; and
a roughness processor configured to determine a roughness metric representative of the topology of the land surface based on at least the ground map and to evaluate the roughness metric for accuracy with respect to the image data, wherein the roughness processor is further configured to
compare the roughness metric with a predetermined roughness threshold, and
release cargo from the work machine on the land surface at the destination when the roughness metric is less than the predetermined roughness threshold.
2. The movable work machine of claim 1 , wherein the roughness processor is further configured to evaluate the roughness metric for accuracy based on the orientation during the traversal of the movable work machine over the one or more obstructions.
3. The movable work machine of claim 2 , wherein the roughness processor is further configured to update the roughness metric using, at least in part, the orientation during the traversal of the movable work machine over the one or more obstructions.
4. The movable work machine of claim 3 , wherein the roughness processor is further configured to determine the roughness metric based on the image data.
5. The movable work machine of claim 1 , wherein the range sensor is one of a LIDAR device or a stereo camera device configured to provide depth assessment.
6. The movable work machine of claim 1 , wherein the roughness processor is further configured to alert an operator when the roughness metric is greater than or equal to the predetermined roughness threshold.
7. A method for verifying terrain at a drop site for a work machine, the method comprising:
receiving, by the work machine, an image signal indicative of image data of land surface at the drop site;
receiving, by the work machine, a range signal indicative of range data of the land surface;
detecting, by the work machine, an obstruction on the land surface from the image data;
based at least in part on the detecting the obstruction, causing the work machine to move over the land surface;
during movement over the land surface, receiving, by the work machine, inertial data indicative of at least pitch and roll of the work machine;
determining a roughness metric for the land surface based at least in part on the pitch and roll of the work machine from the inertial data;
comparing the roughness metric with a predetermined roughness threshold;
determining the roughness metric to be less than the predetermined roughness threshold; and
releasing the payload from the work machine on the land surface at the destination when the roughness metric is less than the predetermined roughness threshold.
8. The method of claim 7 , further comprising:
determining the roughness metric to be greater than or equal to the predetermined roughness threshold;
generating a notification; and
sending the notification to an operator of the work machine for assistance.
9. The method of claim 7 , further comprising:
generating, by a range sensor, the range signal for detecting objects in the environment about the work machine.
10. The method of claim 7 , further comprising:
prior to detecting the obstruction, converting the range data into a ground map of the land surface; and
determining an initial roughness metric based at least in part on the ground map.
11. The method of claim 10 , further comprising:
determining the initial roughness metric to be less than the predetermined roughness threshold; and
releasing cargo from the work machine onto the land surface at the destination.
12. The method of claim 11 , wherein determining the roughness metric comprises updating the initial roughness metric based on the pitch and roll of the work machine from the inertial data.
13. The method of claim 11 , further comprising:
causing the work machine to advance towards the drop site, wherein the receiving the image signal and the receiving the range signal occur during the advance of the work machine towards the drop site.
14. The method of claim 7 , further comprising:
prior to detecting the obstruction, converting the range data into a ground map of the land surface, wherein the roughness metric is based on the ground map and the pitch and roll of the work machine.
15. A semi-autonomous compact loader, comprising:
a chassis;
traction devices coupled between the chassis and ground;
a work tool coupled to a front of the chassis with respect to a direction of forward travel of the semi-autonomous compact loader, the work tool being configured to enable lifting and lowering a payload;
a range sensor configured to generate range data indicative of topology of a land surface at a destination proximate to the semi-autonomous compact loader;
an image sensor configured to generate image data indicative of the topology;
an inertial sensor configured to generate orientation data for the semi-autonomous compact loader during traversal of the land surface;
a terrain mapper configured to convert the range data into a ground map of the land surface;
an obstruction detector configured to identify one or more obstructions on the land surface from image data;
a localization system configured to evaluate orientation of the semi-autonomous compact loader from at least the orientation data; and
a roughness processor configured to determine a roughness metric representative of the topology of the land surface based on at least the ground map and to evaluate the roughness metric for validity with respect to the image data, wherein the roughness processor is further configured to compare the roughness metric with a predetermined roughness threshold and to release the payload from the semi-autonomous compact loader on the land surface at the destination when the roughness metric is less than the predetermined roughness threshold.
16. The semi-autonomous compact loader of claim 15 , wherein the roughness processor is further configured to evaluate the roughness metric for validity based on the orientation during the traversal of the semi-autonomous compact loader over the one or more obstructions.
17. The semi-autonomous compact loader of claim 16 , wherein the roughness processor is further configured to refine the roughness metric based on the orientation during the traversal of the semi-autonomous compact loader over the one or more obstructions.
18. The semi-autonomous compact loader of claim 15 , wherein the image sensor is one of a high dynamic range (HDR) camera and an ultra-sonic camera.
19. The semi-autonomous compact loader of claim 15 , wherein the roughness processor is further configured to alert an operator when the roughness metric is greater than or equal to the predetermined roughness threshold.
20. The semi-autonomous compact loader of claim 15 , wherein the range sensor is one of a LIDAR device or a stereo camera device configured to provide depth assessment.