Detecting negative obstacles
A computer-implemented method when executed by data processing hardware causes the data processing hardware to perform operations. The operations include detecting a candidate support surface at an elevation less than a current surface supporting a legged robot. A determination is made on whether the candidate support surface includes an area of missing terrain data within a portion of an environment surrounding the legged robot, where the area is large enough to receive a touchdown placement for a leg of the legged robot. If missing terrain data is determined, at least a portion of the area of missing terrain data is classified as a no-step region of the candidate support surface. The no-step region indicates a region where the legged robot should avoid touching down a leg of the legged robot.
1 . A computer-implemented method when executed by data processing hardware causes the data processing hardware to perform operations comprising:
identifying a first elevation value for a first elevation of a first portion of an environment that depends at least in part on an orientation of a sensor of a legged robot with respect to the first portion of the environment;
determining, based on sensor data from the sensor, that the first elevation value is less than a second elevation value for a second elevation of a second portion of the environment supporting the legged robot;
determining that at least a subset of the first portion of the environment is occluded from the sensor;
classifying the first portion of the environment as a no-step region indicating a region to avoid touching down a distal end of a leg of the legged robot in response to determining that the first elevation value is less than the second elevation value and determining that the at least a subset of the first portion of the environment is occluded from the sensor; and
instructing movement by the legged robot based on classifying the first portion of the environment as the no-step region.
2 . The method of claim 1 , wherein determining that the first elevation value is less than the second elevation value is based on an output of a terrain mapping system of the legged robot.
3 . The method of claim 1 , wherein classifying the first portion of the environment as the no-step region comprises:
re-classifying the first portion of the environment from a step region to the no-step region.
4 . The method of claim 1 , wherein the first portion of the environment corresponds to a gap in the environment, wherein the operations further comprise:
classifying a third portion of the environment adjacent to the gap as the no-step region.
5 . The method of claim 1 , wherein determining that the first elevation value is less than the second elevation value comprises:
determining that the first portion of the environment corresponds to a gap in the environment, wherein the gap extends from a first position in the environment that is adjacent the legged robot to a second position in the environment.
6 . The method of claim 5 , wherein the operations further comprise:
classifying a third portion of the environment adjacent to the first position as the no-step region; and
classifying a fourth portion of the environment adjacent to the second position as the no-step region.
7 . The method of claim 1 , wherein determining that the first elevation value is less than the second elevation value comprises:
determining that the first portion of the environment corresponds to a gap in the environment, wherein the gap corresponds to a width that is less than a stride length of the legged robot.
8 . The method of claim 1 , wherein instructing movement by the legged robot comprises:
instructing the legged robot to move using four legs of the legged robot, the four legs coupled to a body of the legged robot.
9 . The method of claim 1 , wherein determining that the at least a subset of the first portion of the environment is occluded from the sensor comprises:
determining that the sensor data does not include sensor data associated with the at least a subset of the first portion of the environment.
10 . The method of claim 1 , wherein identifying the first elevation value comprises:
identifying the first elevation value that is different from a third elevation value for a third elevation of a ground surface associated with the first portion of the environment.
11 . The method of claim 1 , wherein the operations further comprise:
obtaining the sensor data from the sensor as disposed on a front end portion of a body of the legged robot.
12 . The method of claim 1 , wherein the operations further comprise:
obtaining the sensor data from the sensor as disposed on a rear end portion of a body of the legged robot.
13 . The method of claim 1 , wherein the operations further comprise:
obtaining the sensor data as the legged robot traverses the environment at a particular speed or gait.
14 . The method of claim 1 , wherein instructing movement by the legged robot comprises:
providing instructions to the legged robot; and
executing, by the legged robot, the instructions, wherein the legged robot moves the distal end of the leg from the second portion of the environment to a third portion of the environment based on executing the instructions.
15 . A robot comprising:
a body;
two or more legs coupled to the body;
a sensor; and
an obstacle detection system comprising data processing hardware and memory hardware in communication with the data processing hardware, the memory hardware storing instructions that when executed on the data processing hardware cause the data processing hardware to:
identify a first elevation value for a first elevation of a first portion of an environment that depends at least in part on an orientation of the sensor with respect to the first portion of the environment;
determine, based on sensor data from the sensor, that the first elevation value is less than a second elevation value for a second elevation of a second portion of the environment supporting at least one of the two or more legs;
determine that at least a subset of the first portion of the environment is occluded from the sensor;
classify the first portion of the environment as a no-step region indicating a region to avoid touching down a distal end of a leg of the two or more legs of the robot in response to determining that the first elevation value is less than the second elevation value and determining that the at least a subset of the first portion of the environment is occluded from the sensor; and
instruct movement by the robot based on classifying the first portion of the environment as the no-step region.
16 . The robot of claim 15 , wherein determining that the first elevation value is less than the second elevation value is based on an output of a terrain mapping system of the robot.
17 . The robot of claim 15 , wherein to classify the first portion of the environment as the no-step region, execution of the instructions on the data processing hardware further causes the data processing hardware to:
reclassify the first portion of the environment from a step region to the no-step region.
18 . The robot of claim 15 , wherein the first portion of the environment corresponds to a gap in the environment, wherein execution of the instructions on the data processing hardware further causes the data processing hardware to:
classify a third portion of the environment adjacent to the gap as the no-step region.
19 . The robot of claim 15 , wherein: the first portion of the environment corresponds to a gap in the environment that extends from a first position in the environment that is adjacent the robot to a second position in the environment.
20 . The robot of claim 19 , wherein execution of the instructions on the data processing hardware further causes the data processing hardware to:
classify a third portion of the environment adjacent to the first position as the no-step region; and
classify a fourth portion of the environment adjacent to the second position as the no-step region.
21 . The robot of claim 15 , wherein the first portion of the environment corresponds to a gap in the environment, wherein the gap corresponds to a width that is less than a stride length of the robot.
22 . The robot of claim 15 , wherein the two or more legs comprise four legs.
23 . The robot of claim 15 , wherein to determine that the at least a subset of the first portion of the environment is occluded from the sensor, execution of the instructions on the data processing hardware further causes the data processing hardware to:
determine that the sensor data does not include sensor data associated with the at least a subset of the first portion of the environment.
24 . The robot of claim 15 , wherein a third elevation value for a third elevation of a ground surface associated with the first portion of the environment is different from the first elevation value and the second elevation value.
25 . The robot of claim 15 , wherein the sensor is disposed on a front end portion of the body.
26 . The robot of claim 15 , wherein the sensor is disposed on a rear end portion of the body.
27 . The robot of claim 15 , wherein the robot traverses the environment at a particular speed or gait.
28 . The robot of claim 15 , wherein to instruct movement by the robot, execution of the instructions on the data processing hardware further causes the data processing hardware to:
provide a set of instructions to the robot; and
execute, by the robot, the set of instructions, wherein the robot moves the distal end of the leg from the second portion of the environment to a third portion of the environment based on executing the instructions.