Remote physiological data sensing robot
Techniques described in this application are directed to an autonomous robot that is configured to interact with its environment through identifying the presence and emotional state (e.g., mood) of live subjects in the scene. The autonomous robot includes one or more remote physiological data collecting sensors capable of remotely collecting physiological data such as heart rate, blood circulation, or respiratory activity of the live subjects within close proximity of the robot.
1 . A robot comprising:
one or more sensors configured to collect sensor data by remotely measuring one or more physiological functions of at least one live subject;
one or more output devices;
one or more processors; and
a non-transitory computer-readable medium including instructions that, when executed by the one or more processors, cause the one or more processors to:
identify the at least one live subject within a predetermined distance of the robot;
receive, from the one or more sensors, the sensor data associated with the one or more physiological functions of the at least one live subject;
determine an aggressive emotional state of the at least one live subject using at least the sensor data;
determine a robot action based on the aggressive emotional state of the at least one live subject;
provide, to the one or more output devices, an instruction to initiate the robot action; and
perform the robot action using the one or more output devices, wherein the robot action includes activating a vehicle system of the robot to move away from the at least one live subject upon determining, based at least in part on the aggressive emotional state of the at least one live subject, a threshold likelihood of the at least one live subject attacking the robot.
2 . The robot of claim 1 , wherein the aggressive emotional state of the at least one live subject and the robot action are determined using a machine learning model.
3 . The robot of claim 1 , wherein the robot action further includes one or more of activating a series of lights to imitate a human expression on a user interface displayed on the one or more output devices, contacting a remote operator, and playing an audio file.
4 . The robot of claim 1 , further comprising:
one or more video cameras,
wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:
capture image data of the at least one live subject using the one or more video cameras;
correlate the sensor data with the image data; and
determine the aggressive emotional state of the at least one live subject using correlation between the sensor data and the image data.
5 . The robot of claim 4 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:
determine the at least one live subject is a person or an animal using the image data.
6 . The robot of claim 1 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:
after performing the robot action by the one or more output devices,
continue monitoring the at least one live subject by receiving a stream of sensor data at predetermined intervals;
determine a change in the sensor data associated with one or more physiological functions of the at least one live subject;
determine a change in the aggressive emotional state of the at least one live subject based on the change in the sensor data;
determine a subsequent robot action based on the change in the aggressive emotional state of the at least one live subject;
provide, to the one or more output devices, an instruction to initiate the subsequent robot action; and
perform the subsequent robot action using the one or more output devices.
7 . The robot of claim 1 , wherein receiving sensor data includes receiving a stream of sensor data at predetermined intervals during an interaction between the robot and the at least one live subject.
8 . The robot of claim 1 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:
determine a confidence score associated with the robot action; and
determine that the confidence score exceeds a threshold value prior to performing the robot action.
9 . The robot of claim 1 , wherein the one or more physiological functions include one or more of a cardiac activity including a heart rate, a respiratory activity, or a body temperature.
10 . The robot of claim 1 , wherein the one or more sensors include:
a first sensor that uses microwave signals; and
a second sensor that uses one of Doppler signals, ultra-wideband signals, or electromagnetic radio-frequency (RF) signals to remotely measure the one or more physiological functions of the at least one live subject.
11 . A computer-implemented method comprising:
identifying, by a robot, at least one live subject within a predetermined distance of the robot;
collecting, by the robot, sensor data by remotely measuring one or more physiological functions of the at least one live subject, wherein the at least one live subject is a person or an animal;
determining, by the robot, an aggressive emotional state of the at least one live subject using at least the sensor data;
determining, by the robot, a robot action based on the aggressive emotional state of the at least one live subject;
providing, by the robot to one or more output devices of the robot, an instruction to initiate the robot action; and
performing, by the robot, the robot action using the one or more output devices,
wherein upon determining, based at least in part on the aggressive emotional state of the at least one live subject, a threshold likelihood of the at least one live subject attacking the robot, the robot performs the robot action of activating a vehicle system of the robot to move away from the at least one live subject.
12 . The computer-implemented method of claim 11 , wherein the aggressive emotional state of the at least one live subject and the robot action are determined using a machine learning model.
13 . The computer-implemented method of claim 11 , wherein the robot action further includes one or more of activating a series of lights to imitate a human expression on a user interface displayed on the one or more output devices, contacting a remote operator, and playing an audio file.
14 . The computer-implemented method of claim 11 , further comprising:
capturing, by one or more video cameras of the robot, image data of the at least one live subject;
correlating the sensor data with the image data; and
determining the aggressive emotional state of the at least one live subject using correlation between the sensor data and the image data.
15 . The computer-implemented method of claim 14 , further comprising:
determining the at least one live subject is a person or an animal using the image data.
16 . The computer-implemented method of claim 11 , further comprising:
after performing the robot action by the one or more output devices,
continuing monitoring the at least one live subject by receiving a stream of sensor data at predetermined intervals;
determining a change in the sensor data associated with one or more physiological functions of the at least one live subject;
determining a change in the aggressive emotional state of the at least one live subject based on the change in the sensor data;
determining a subsequent robot action based on the change in the aggressive emotional state of the at least one live subject;
providing, to the one or more output devices, an instruction to initiate the subsequent robot action; and
performing the subsequent robot action using the one or more output devices.
17 . The computer-implemented method of claim 11 , wherein receiving sensor data includes receiving a stream of sensor data at predetermined intervals during an interaction between the robot and the at least one live subject.
18 . The computer-implemented method of claim 11 , further comprising:
determining a confidence score associated with the robot action; and
determining that the confidence score exceeds a threshold value prior to performing the robot action.
19 . The computer-implemented method of claim 11 , wherein the one or more physiological functions include one or more of a cardiac activity including a heart rate, a respiratory activity, or a body temperature.
20 . The computer-implemented method of claim 11 , wherein collecting the sensor data includes using microwave signals, Doppler signals, ultra-wideband signals, or electromagnetic radio-frequency (RF) signals to remotely measure the one or more physiological functions of the at least one live subject.