Systems and methods for robotic path planning
Systems and methods for robotic path planning are disclosed. In some implementations of the present disclosure, a robot can generate a cost map associated with an environment of the robot. The cost map can comprise a plurality of pixels each corresponding to a location in the environment, where each pixel can have an associated cost. The robot can further generate a plurality of masks having projected path portions for the travel of the robot within the environment, where each mask comprises a plurality of mask pixels that correspond to locations in the environment. The robot can then determine a mask cost associated with each mask based at least in part on the cost map and select a mask based at least in part on the mask cost. Based on the projected path portions within the selected mask, the robot can navigate a space.
1. A method for determining a trajectory of a robot, comprising:
generating, via a controller of the robot, a cost map corresponding to an environment, the cost map comprises a computer readable map of the environment produced using data from sensors coupled to the robot, wherein the cost map comprises a cost function as a function space to assign a cost for each pixel of the computer readable map, and the cost for each pixel is based at least in part on (i) presence of objects or lack thereof, and (ii) distance from a pre-determined route;
receiving, via the controller, a matrix comprising a plurality of pre-determined masks, each pre-determined mask of the plurality of pre-determined masks corresponding to an associated actuator command of the robot;
evaluating, via the controller, a cost for each pre-determined mask of the plurality of pre-determined masks based on the cost map;
identifying, via the controller, a pre-determined mask from the plurality of pre-determined masks, the pre-determined mask comprising the cost with a lowest value; and
executing the actuator command associated with the identified pre-determined mask of the lowest value.
2. The method of claim 1 , wherein, the pre-determined route corresponds to at least one of (i) a shortest path from one location to another, or (ii) a route previously demonstrated to or learned by the robot.
3. The method of claim 1 , further comprising:
evaluating, via the controller, the cost for each pre-determined mask of the matrix based on an area occupied by each mask and cumulative cost of pixels of the cost map within the area, the area occupied corresponding to an area occupied by the robot during execution of the actuator command associated with the pre-determined mask.
4. The method of claim 3 , wherein the cost map and the matrix of pre-determined masks comprise images.
5. The method of claim 1 , further comprising:
increasing, via the controller, a cost of pixels corresponding to regions where the robot has previously navigated, the increase in the cost of pixels corresponds to a preference for the robot to not travel over an area multiple times.
6. A robot, comprising:
a non-transitory computer readable medium comprising a plurality of computer readable instructions which, when executed by a controller, cause the controller to
generate a cost map corresponding to an environment, the cost map comprises a computer readable map of the environment produced using data from sensors coupled to a robot, wherein the cost map comprises a cost function as a function space to assign a cost for each pixel of the computer readable map, and the cost for each pixel is based at least in part on (i) a presence of objects or lack thereof, and (ii) distance from a pre-determined route;
receive a matrix comprising a plurality of pre-determined masks, each pre-determined mask of the plurality of pre-determined masks corresponding to an associated actuator command of the robot;
evaluate a cost for each pre-determined mask of the plurality of pre-determined masks based on the cost map;
identify a pre-determined mask from the plurality of pre-determined masks, the pre-determined mask comprising the cost with a lowest value; and
execute the actuator command associated with the identified pre-determined mask of the lowest value.
7. The robot of claim 6 , wherein, the pre-determined route corresponds to at least one of (i) a shortest path from one location to another, or (ii) a route previously demonstrated to or learned by the robot.
8. The robot of claim 6 , wherein the controller is further configured to execute the computer readable instructions to,
evaluate the cost for each pre-determined mask of the matrix based on an area occupied by each mask and the cumulative cost of pixels of the cost map within the area, the area occupied corresponding to an area occupied by the robot during execution of the actuator command associated with the pre-determined mask.
9. The robot of claim 8 , wherein the cost map and the matrix of pre-determined masks comprise images.
10. The robot of claim 6 , wherein the controller is further configured to execute the computer readable instructions to
increase a cost of pixels corresponding to regions where the robot has previously navigated, the increase in the cost of pixels corresponds to a preference for the robot to not travel over an area multiple times.
11. A non-transitory computer readable medium comprising a plurality of computer readable instructions which, when executed by a controller of, cause the controller to
generate a cost map corresponding to an environment, the cost map comprises a computer readable map of the environment produced using data from sensors coupled to a robot, wherein the cost map comprises a cost function as a function space to assign a cost for each pixel of the computer readable map, and the cost for each pixel is based at least in part on (i) presence of objects or lack thereof, and (ii) distance from a pre-determined route;
receive a matrix comprising a plurality of pre-determined masks, each pre-determined mask of the plurality of pre-determined masks corresponding to an associated actuator command of the robot;
evaluate a cost for each pre-determined mask of the plurality of pre-determined masks based on the cost map;
identify a pre-determined mask from the plurality of pre-determined masks, the pre-determined mask comprising the cost with a lowest value; and
execute the actuator command associated with the identified pre-determined mask of the lowest value.
12. The non-transitory computer readable medium of claim 11 , wherein
the pre-determined route corresponds to at least one of (i) a shortest path from one location to another, or (ii) a route previously demonstrated to or learned by the robot.
13. The non-transitory computer readable medium of claim 11 , wherein the controller is further configured to execute the computer readable instructions to
evaluate the cost for each pre-determined mask of the matrix based on an area occupied by each mask and cumulative cost of pixels of the cost map within the area, the area occupied corresponding to an area occupied by the robot during execution of the actuator command associated with the pre-determined mask.
14. The non-transitory computer readable medium of claim 13 , wherein the cost map and the matrix of pre-determined masks comprise of images.
15. The non-transitory computer readable medium of claim 11 , wherein the controller is further configured to execute the computer readable instructions to
increase a cost of pixels corresponding to regions where the robot has previously navigated, the increase in the cost of pixels corresponds to a preference for the robot to not travel over an area multiple times.