Robotic social interaction
A robot may identify a human located proximate to the robot in a physical environment based on sensor data captured from one or more sensors on the robot. A trajectory of the human through space may be predicted. When the predicted trajectory of the human intersects with a current path of the robot, an updated path to a destination location in the environment may be determined so as to avoid a collision between the robot and the human along the predicted trajectory. The robot may then move along the determined path.
1. A method implemented at a robot, the method comprising:
autonomously traversing a physical environment along a first path from a first location to a second location via a mobility apparatus, the first path determined based on sensor data collected from a sensor module at the robot, the sensor data characterizing the physical environment;
autonomously identifying via a processor a human located within the physical environment;
identifying a role associated with the human;
determining a communication mode for communicating with the human based on the identified role;
autonomously determining via the processor a second path through the physical environment, the human being predicted to traverse the second path in the future based on the identified role;
autonomously determining a first one or more future predicted distances between the human and the robot were the robot to follow the first path and were the human to follow the second path;
autonomously determining via the processor that a first predicted distance of the first one or more future predicted distances falls below a designated threshold;
autonomously determining via the processor a third path through the physical environment, the third path ending at the second location, the third path being determined based on the second path so that a second one or more future predicted distances between the human and the robot were the robot to follow the first path and were the human to follow the second path each exceed the designated threshold;
autonomously updating the traversal of the physical environment from the first path to the third path; and
presenting a visual cue indicating a navigational goal for the robot based on the third path and the communication mode.
2. The method recited in claim 1 , the method further comprising:
determining a model of the physical environment based on the sensor data, the model representing one or more surfaces within the physical environment.
3. The method recited in claim 2 , wherein the model is a three-dimensional model, the three-dimensional model representing one or more stationary physical objects, the first and third paths being determined to avoid the one or more stationary physical objects.
4. The method recited in claim 1 , wherein predicting the second path comprises identifying a first plurality of locations, each of the first plurality of locations being associated with a first respective point in time, the human being present at each of the first plurality of locations at the first respective point in time.
5. The method recited in claim 4 , wherein predicting the second path further comprises determining a second predicted plurality of locations, each of the second predicted plurality of locations being associated with a second respective point in time, the human being predicted to be present at each of the second plurality of locations at the second respective point in time.
6. The method recited in claim 1 , wherein the robot includes a communication interface facilitating autonomous communication with the human.
7. The method recited in claim 1 , wherein the sensor module includes a camera, and wherein the sensor data includes image data, and wherein the human is identified at least in part by analyzing the image data via the processor.
8. The method recited in claim 1 , wherein the sensor module includes a depth sensor, and wherein the sensor data includes depth information, and wherein predicting the second path involves analyzing the depth information.
9. A robot comprising:
a sensor module configured to collect sensor data characterizing a physical environment;
a mobility apparatus configured to traverse the physical environment; and
a processor configured to:
determine a first path from a first location in the physical environment to a second location in the physical environment;
instruct the mobility apparatus to traverse the physical environment along the first path;
identify a human located within the physical environment;
identify a role associated with the human;
determine a communication mode for communicating with the human based on the identified role;
determine a second path through the physical environment, the human being predicted to traverse the second path in the future based on the identified role;
determine a first one or more future predicted distances between the human and the robot were the robot to follow the first path and were the human to follow the second path;
determine that a first predicted distance of the first one or more future predicted distances falls below a designated threshold;
determine via the processor a third path through the physical environment, the third path ending at the second location, the third path being determined based on the second path so that a second one or more future predicted distances between the human and the robot were the robot to follow the first path and were the human to follow the second path each exceed the designated threshold;
instruct the mobility apparatus to update the traversal of the physical environment from the first path to the third path; and
present a visual cue indicating a navigational goal for the robot based on the third path and the communication mode.
10. The robot recited in claim 9 , wherein the processor is further operable to:
determine a three-dimensional model of the physical environment based on the sensor data, the three-dimensional model representing one or more surfaces within the physical environment, the three-dimensional model further representing one or more stationary physical objects, the first and third paths being determined to avoid the one or more stationary physical objects.
11. The robot recited in claim 9 , wherein predicting the second path comprises identifying a first plurality of locations, each of the first plurality of locations being associated with a first respective point in time, the human being present at each of the first plurality of locations at the first respective point in time, and wherein predicting the second path further comprises determining a second predicted plurality of locations, each of the second predicted plurality of locations being associated with a second respective point in time, the human being predicted to be present at each of the second plurality of locations at the second respective point in time.
12. One or more non-transitory machine readable media having instructions stored thereon for performing a method, the method comprising:
autonomously traversing a physical environment along a first path from a first location to a second location via a mobility apparatus, the first path determined based on sensor data collected from a sensor module at the robot, the sensor data characterizing the physical environment;
autonomously identifying via a processor a human located within the physical environment;
identifying a role associated with the human;
determining a communication mode for communicating with the human based on the identified role;
autonomously determining via the processor a second path through the physical environment, the human being predicted to traverse the second path in the future based on the identified role;
autonomously determining a first one or more future predicted distances between the human and the robot were the robot to follow the first path and were the human to follow the second path;
autonomously determining via the processor that a first predicted distance of the first one or more future predicted distances falls below a designated threshold;
autonomously determining via the processor a third path through the physical environment, the third path ending at the second location, the third path being determined based on the second path so that a second one or more future predicted distances between the human and the robot were the robot to follow the first path and were the human to follow the second path each exceed the designated threshold;
autonomously updating the traversal of the physical environment from the first path to the third path; and
presenting a visual cue indicating a navigational goal for the robot based on the third path and the communication mode.