Mitigating sensor noise in legged robots
An example implementation involves receiving measurements from an inertial sensor coupled to the robot and detecting an occurrence of a foot of the legged robot making contact with a surface. The implementation also involves reducing a gain value of an amplifier from a nominal value to a reduced value upon detecting the occurrence. The amplifier receives the measurements from the inertial sensor and provides a modulated output based on the gain value. The implementation further involves increasing the gain value from the reduced value to the nominal value over a predetermined duration of time after detecting the occurrence. The gain value is increased according to a profile indicative of a manner in which to increase the gain value of the predetermined duration of time. The implementation also involves controlling at least one actuator of the legged robot based on the modulated output during the predetermined duration of time.
1. A method comprising:
receiving, at a control system of legged robot, sensor data from an inertial sensor disposed on a leg of the legged robot;
determining, by the control system, that the sensor data comprises a noise signal that satisfies a noise threshold, the noise threshold indicating a level of noise that adversely impacts the control system;
generating, by the control system, a time-varying gain signal based on the noise signal of the sensor data, the time-varying gain signal suppressing the noise signal of the sensor data; and
controlling, by the control system, at least one actuator on the leg of the legged robot using a modulated output from an amplifier, the modulated output from the amplifier generated with the time-varying gain signal.
2. The method of claim 1 , further comprising detecting, by the control system, that a foot of the leg of the legged robot contacts a surface within an environment of the legged robot.
3. The method of claim 1 , wherein the noise signal corresponds to vibrations from contact between the leg of the legged robot and a surface within an environment of the legged robot.
4. The method of claim 3 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal changing from the reduced gain value to the nominal gain value over a predetermined duration of time, the predetermined duration of time based on a decay rate of the vibrations from the contact between the leg of the legged robot and a surface within an environment of the legged robot.
5. The method of claim 1 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal linearly increasing from the reduced gain value to the nominal gain value over a predetermined duration of time.
6. The method of claim 1 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal exponentially increasing from the reduced gain value to the nominal gain value over a predetermined duration of time.
7. The method of claim 1 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal changing from the reduced gain value to the nominal gain value over a predetermined duration of time, the predetermined duration of time based on a stance duration for the leg of the legged robot.
8. The method of claim 1 , wherein the sensor data comprises an acceleration measurement and the noise signal corresponds to an oscillating value of the acceleration measurement.
9. The method of claim 8 , further comprising detecting, by the control system, that the leg of the legged robot experiences an acceleration change based on the acceleration measurement, the acceleration change defined by a change from a positive acceleration measurement or a negative acceleration measurement to an approximately zero measurement of acceleration.
10. The method of claim 1 , further comprising filtering, by the control system, the sensor data prior to determining that the sensor data comprises the noise signal that satisfies the noise threshold.
11. A robot comprising:
a body;
two or more legs coupled to the body and configured to move the robot about an environment of the robot, each leg of the two or more legs comprising at least one actuator;
an inertial sensor disposed on at least one leg of the two or more legs; and
a control system of the robot configured to perform operations comprising:
receiving sensor data from the inertial sensor;
determining that the sensor data comprises a noise signal that satisfies a noise threshold, the noise threshold indicating a level of noise that adversely impacts the control system;
generating a time-varying gain signal based on the noise signal of the sensor data, the time-varying gain signal suppressing the noise signal of the sensor data; and
controlling the at least one actuator of a respective leg of the robot using a modulated output from an amplifier, the modulated output from the amplifier generated with the time-varying gain signal.
12. The robot of claim 11 , wherein the operations further comprise detecting that a foot of the respective leg of the robot contacts a surface within the environment of the robot.
13. The robot of claim 11 , wherein the noise signal corresponds to vibrations from contact between the respective leg of the robot and a surface within an environment of the robot.
14. The robot of claim 13 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal changing from the reduced gain value to the nominal gain value over a predetermined duration of time, the predetermined duration of time based on a decay rate of the vibrations from the contact between the respective leg of the robot and a surface within the environment of the robot.
15. The robot of claim 11 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal linearly increasing from the reduced gain value to the nominal gain value over a predetermined duration of time.
16. The robot of claim 11 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal exponentially increasing from the reduced gain value to the nominal gain value over a predetermined duration of time.
17. The robot of claim 11 , wherein the time-varying gain signal comprises a nominal gain value and reduced gain value with respect to the nominal gain value, the time-varying gain signal changing from the reduced gain value to the nominal gain value over a predetermined duration of time, the predetermined duration of time based on a stance duration for the respective leg of the robot.
18. The robot of claim 11 , wherein the sensor data comprises an acceleration measurement and the noise signal corresponds to an oscillating value of the acceleration measurement.
19. The robot of claim 18 , wherein the operations further comprise detecting that the respective leg of the robot experiences an acceleration change based on the acceleration measurement, the acceleration change defined by a change from a positive acceleration measurement or a negative acceleration measurement to an approximately zero measurement of acceleration.
20. The robot of claim 11 , wherein the operations further comprise filtering the sensor data prior to determining that the sensor data comprises the noise signal that satisfies the noise threshold.