Mechanical arm control device and method

Abstract

A robotic arm control device is configured to control a robotic arm module. The robotic arm control device includes an information acquisition module and a control module. The information acquisition module includes an electrostatic sensor configured to generate a potential signal. The control module is coupled to the information acquisition module and includes a memory, a state recognition module, and a robotic arm control module. The memory is configured to store state values. The state recognition module is configured to capture signal segments contained in the potential signal and execute a recognition model configured to read and classify the signal segments to update the state values ​​according to the category of the signal segments. The robotic arm control module is configured to read the state values ​​and control the robotic arm module accordingly.

Claims

1. A robotic arm control device configured to control a robotic arm module, the robotic arm control device comprising: an information acquisition module including at least one electrostatic sensor configured to generate a potential signal; and a control module coupled to the information acquisition module, the control module comprising: a memory configured to store a state value; and a state recognition module configured to: capture a signal segment included in the potential signal; and execute a recognition model configured to read and classify the signal segment to update the state value according to the category of the signal segment, wherein... The recognition model is configured to read the signal segment to generate multiple status signals, and the status recognition module is configured to calculate an area overlap rate between one of the multiple status signals and a template signal in a time sequence to classify the signal segment; and a robotic arm control module is configured to read the status value and control the robotic arm module accordingly.

2. The robotic arm control device as described in claim 1, wherein, The information acquisition module includes multiple electrostatic sensors configured to be worn on multiple fingers of the same hand.

3. The robotic arm control device as described in claim 2, wherein, The information acquisition module includes a hand electrode device and an analog-to-digital converter. The hand electrode device includes a reference electrode and a plurality of finger rings, each of which is an electrostatic sensor configured to be fitted onto the finger. The reference electrode is configured to be attached to the back of the palm. The analog-to-digital converter is configured to read the reference electrode and the plurality of finger rings to generate the potential signal.

4. The robotic arm control device as described in claim 2, wherein, The information acquisition module includes multiple rings, each of which is an electrostatic sensor. Each ring includes: an annular contact layer containing multiple conical protrusions arranged in an array on the inner surface of the annular contact layer; an annular electrode layer attached to the outer surface of the annular contact layer; an annular housing attached to the outer surface of the annular electrode layer and having a through-hole; and a wire passing through the through-hole of the annular housing to be electrically connected to the annular electrode layer.

5. The robotic arm control device as described in claim 1, wherein, The status recognition module is configured to scan the potential signal according to a moving pane, and in response to determining that the moving pane contains an absolute value higher than a potential threshold, the potential signal within the moving pane is captured as the signal segment.

6. The robotic arm control device as described in claim 1, wherein, The status recognition module is configured to: determine a time range in which the area overlap rate is greater than an overlap rate threshold; and set the value of a status signal outside the time range to 0.

7. The robotic arm control device as described in claim 6, wherein the recognition model classifies the signal segment into the category corresponding to one of its state signals in response to determining that the area overlap rate contains a value greater than the overlap rate threshold.

8. The robotic arm control device as described in claim 6, wherein, The state recognition module detects the peak time point of one of its state signals within the time range to generate a state transition point, and updates the state value at the state transition point.

9. The robotic arm control device as described in claim 1, wherein, The information acquisition module includes multiple electrostatic sensors. The state recognition module is configured to capture the potential signals generated by each of the multiple electrostatic sensors to generate multiple signal segments. The recognition model is configured to read the multiple signal segments to classify one of the multiple signal segments and update the state value according to the category of one of the signal segments.

10. The robotic arm control device as described in claim 9, wherein, The memory is configured to store a first state value and a second state value. The recognition model is configured to update the first state value according to the category of a signal segment. The robotic arm control module controls the robotic arm module to move along a first degree of freedom according to the first state value and controls the robotic arm module to move along a second degree of freedom according to the second state value.

11. The robotic arm control device as described in claim 9, wherein, The memory is configured to store a first state value, the recognition model is configured to update the first state value according to the category of a signal segment thereof, and the robotic arm control module is configured to: in response to the first state value corresponding to a first state category, manipulate the robotic arm module to perform a first action along a first degree of freedom; and in response to the first state value corresponding to a second state category, manipulate the robotic arm module to perform a second action along the first degree of freedom.

12. The robotic arm control device as described in claim 11, wherein, The vectors of the first action and the second action are in opposite directions.

13. A method for controlling a robotic arm, comprising: receiving a potential signal; capturing a signal segment contained in the potential signal; executing a recognition model configured to read the signal segment to generate multiple state signals, and the state recognition module configured to calculate an area overlap rate in time between one of the multiple state signals and a template signal to classify the signal segment; updating a state value according to the category of the signal segment; and controlling a robotic arm module according to the state value.

14. The robotic arm control method as described in claim 13 further comprises: receiving a first potential signal and a second potential signal; capturing a first signal segment included in the first potential signal and a second signal segment included in the second potential signal; executing the recognition model to read the first signal segment and the second signal segment to classify the first signal segment; and updating the state value according to the category of the first signal segment.

15. The robotic arm control method as described in claim 13 further includes: in response to the state value corresponding to a first state category, controlling the robotic arm module to perform a first action along a first degree of freedom; and in response to the state value corresponding to a second state category, controlling the robotic arm module to perform a second action along the first degree of freedom, wherein the vector directions of the first action and the second action are opposite.

16. The robotic arm control method as described in claim 13 further comprises: reading a signal segment training set, the signal segment training set comprising a plurality of signal segment groups, each signal segment group comprising at least one signal segment; reading a state signal training set, the state signal training set comprising a plurality of state signal groups, each state signal group comprising a first state signal and at least one second state signal, the period of the first state signal comprising a probability peak, and the period of the at least one second state signal comprising a probability of zero; inputting one of the signal segment groups into the recognition model, the one of the signal segment groups corresponding to one of the state signal groups; and obtaining the minimum value of the loss function between the output of the recognition model and the one of the state signal groups, to complete the training of the recognition model.

17. A robotic arm control device configured to control a robotic arm module, the robotic arm control device comprising: an information acquisition module including a hand electrode device and an analog-to-digital converter, the hand electrode device including a reference electrode and a plurality of finger rings, each finger ring being an electrostatic sensor, the analog-to-digital converter being configured to read the reference electrode and the plurality of finger rings to generate a potential signal; and a control module coupled to the information acquisition module, the control module comprising: a memory configured to store a state value; a state recognition module configured to: scan the potential signal according to a moving pane; update the state value to a first state category in response to determining that the moving pane contains a value higher than or equal to a first potential threshold; and update the state value to a second state category in response to determining that the moving pane contains a value lower than the first potential threshold and higher than or equal to a second potential threshold; and a robotic arm control module configured to read the state value and control the robotic arm module accordingly.

18. The robotic arm control device as described in claim 17, wherein, The control module further includes a state value maintained in response to determining that the moving pane contains a positive value below the second potential threshold. The robotic arm control module is configured to continuously operate the robotic arm module according to the state value.

19. The robotic arm control device as described in claim 17, wherein, The robotic arm control module is configured to: in response to the state value corresponding to the first state category, manipulate the robotic arm module to perform a first action along a first degree of freedom; and in response to the state value corresponding to the second state category, manipulate the robotic arm module to perform a second action along the first degree of freedom, wherein the vector directions of the first action and the second action are opposite.

20. The robotic arm control device as described in claim 19, wherein, The control module responds to the state value corresponding to the first state category and determines that the moving pane contains a negative value below a third potential threshold, updates the state value to a third state category, and the robotic arm control module responds to the state value corresponding to the third state category by controlling the robotic arm module to stop the movement along the first degree of freedom, wherein the third potential threshold has the opposite polarity to the first potential threshold.

Images