Characterizing performance of a robot joint

Abstract

A method for characterizing the performance of a joint in a surgical robotic arm, the joint being driven by a transmission system that transmits power from a drive source to the joint, the method comprising: sending a first command signal to position the robotic arm in an initial configuration; sending a second command signal to apply a force to the joint, thereby displacing the joint from a stable state; for a plurality of predefined time intervals: receiving a first measurement indicating the configuration of the drive source at a first end of the transmission system; receiving a second measurement indicating the configuration of the joint at a second end of the transmission system; calculating an elongation value using the first and second measurements; and receiving a third measurement indicating a torque experienced by the joint at the second end of the transmission system; comparing the elongation value with a corresponding torque value at each of the predefined time intervals; and generating an output indicating the performance of the joint from the comparison.

Description

Technical Field

[0001] This invention relates to a test for characterizing the performance of joints in a surgical robot arm. Background Technology

[0002] A typical robot manipulator comprises a series of rigid elements connected together by one or more joints, which can be engaged in series to form an arm. The manipulator also includes a base unit located at a first end of the series of rigid elements and an end effector located at a second end opposite the first end. Each joint of the manipulator is driven by a drive source and a corresponding transmission system, thereby causing relative movement of the rigid elements. This relative movement is used to change the configuration of the end effector in a desired position. Each joint can provide rotational and / or linear motion. The drive source can be any suitable component, such as an electric motor or a hydraulic actuator.

[0003] The transmission system associated with a given joint typically includes a gearbox. The gearbox can alter the forces applied to the joint from the drive source. A typical gearbox includes gears or toothed components that allow sliding within the transmission system. The use of toothed components introduces backlash, which ultimately leads to motion losses within the transmission system under certain joint configurations. These motion losses and their impact on joint configurations are difficult to predict. Therefore, it is necessary to test and obtain real-world data indicating the joint's performance in a range of different configurations. Understanding the joint's performance and its corresponding drive mechanism is beneficial for obtaining parameters that allow for comparison and determination of the joint's performance during use, and for adjusting the control parameters supplied to the drive source.

[0004] There are many known tests for characterizing the performance of robot joints. These known tests typically involve the use of external testing equipment and sensors. For example, a test may include locking one end of a drivetrain while forcing the other end to rotate in order to measure the torque and configuration of the joint at the unlocked end of the drivetrain. Alternatively, or in addition to these, known tests involve manually applying forces to the joints in the robotic arm and monitoring changes in the input and output joint configurations as that torque is applied. A drawback associated with using external equipment and sensors to characterize joint performance is that it introduces complexity and reliability issues into the testing process. Furthermore, known testing methods cannot observe the entire range of joint configurations and forces. Therefore, it is impossible to obtain all the data points needed for a comprehensive investigation of joint performance.

[0005] There is a need for an automated, fast, reliable, and standardized test to characterize the performance of robot joints. Summary of the Invention

[0006] According to a first aspect, a method is provided for characterizing the performance of a joint in a surgical robotic arm, the joint being driven by a transmission system that transmits power from a drive source to the joint, the method comprising: sending a first command signal to position the robotic arm in an initial configuration; sending a second command signal to apply a force to the joint, thereby displacing the joint from a stable state; for a plurality of predefined time intervals: receiving a first measurement indicating the configuration of the drive source at a first end of the transmission system; receiving a second measurement indicating the configuration of the joint at a second end of the transmission system; calculating an elongation value using the first measurement and the second measurement; and receiving a third measurement indicating a torque experienced by the joint at the second end of the transmission system; comparing the elongation value with a corresponding torque value at each of the predefined time intervals; and generating an output indicating the performance of the joint from the comparison.

[0007] Generating the output may include calculating a graph of elongation versus torque to define the stiffness characteristics of the joint.

[0008] The method may further include comparing the elongation value and the corresponding torque value with predetermined characteristics of the joint.

[0009] The method may further include sending a command signal to apply an instantaneous force to displace the joint from the stable state; and receiving a first measurement, a second measurement, and a third measurement at multiple predefined time intervals when the joint returns to the stable state.

[0010] The method may further include sending a command signal to consistently increase the frequency of the force applied to the joint; and receiving the first measurement, the second measurement, and the third measurement at multiple predefined time intervals.

[0011] The method may further include sending a command signal to consistently change the direction of the force to be applied to the joint; and receiving the first measurement, the second measurement, and the third measurement at multiple predefined time intervals.

[0012] The second command signal can indicate the desired configuration of the joint.

[0013] Elongation can be characterized by the measurement configuration q of the drive source. i Measurement configuration q of the joint o The differences between them.

[0014] The method may also include calculating one or more regression lines on a graph of elongation versus torque, each regression line having a corresponding gradient.

[0015] The method may also include analyzing the gradient of the one or more regression lines to identify multiple distinct regions.

[0016] The method may further include identifying recoil regions from the graph, the recoil regions being characterized by regression lines having a gradient close to zero.

[0017] The method may further include identifying one or more linear stiffness regions from the graph, each linear stiffness region being identified by a regression line having a gradient above a predetermined threshold.

[0018] The method may also include analyzing the graph to determine the accuracy of the one or more regression lines.

[0019] Analyzing the graph may include performing R-squared analysis.

[0020] The multiple different regions can be identified by applying signal conditioning to the received sensor measurements.

[0021] The driving source can be an electric motor.

[0022] The pre-defined configuration of the arm can be selected from a set of pre-defined arm configurations.

[0023] A transmission system may include one or more gears.

[0024] The transmission system can be a harmonic drive. Detailed Implementation

[0025] The invention will now be described by way of example with reference to the accompanying drawings. In the drawings:

[0026] Figure 1 The configuration of the robotic arm is shown;

[0027] Figure 2 A system for controlling the configuration of joints in a robotic arm is shown;

[0028] Figure 3 The characteristics of the nonlinear stiffness of the transmission system of a robot joint are shown.

[0029] Figure 4 This demonstrates a method for characterizing the performance of joints in a surgical robotic arm.

[0030] The following description describes the technology in the context of robotic systems. Robotic systems may include manufacturing systems, such as vehicle manufacturing systems, parts handling systems, laboratory systems, and manipulators, such as manipulators for hazardous materials or surgical manipulators. Figure 1The robotic system shown is specifically a surgical robotic system. However, the features described below are not limited to this type of system, but are more generally applicable to robotic systems.

[0031] Figure 1 A surgical robot with an arm 100 extending from a base unit 102 is shown. The arm includes multiple rigid limbs 104a-e connected by multiple rotary joints 106a-e. The rotary joints 106a-e are configured to apply rotational motion. Instead of or in addition to these rotary joints, the surgical robot may include one or more joints for providing linear motion. The limb closest to the base unit 102 is the proximal limb 104a and is connected to the base unit via a proximal joint 106a. The remaining limbs of the arm are each connected in series via joints in multiple joints 106b-e. A wrist 108 may include four individual rotary joints. The wrist 108 connects a limb (104d) to the distal limb (104e) of the arm. The distal limb 104e carries an attachment 110 of a surgical instrument 112. Each joint 106a-e of the arm 100 has one or more drive sources 114 that can be operated to induce rotational motion at the corresponding joint. Each drive source 114 is connected to its corresponding joint 106a-e via a transmission system that transfers power from the drive source to the joint. In one example, the transmission system includes one or more gears. In a more specific example, the transmission system is a harmonic drive. The harmonic drive includes a wave generator, a concentric arrangement of flexible splines and circular splines, and increases the torque that can be applied to the joint by the drive source.

[0032] In one example, the drive source 114 is an electric motor. Alternatively, the drive source 114 may be a hydraulic actuator or any other suitable component. Each joint 106a-e also includes one or more configuration and / or force sensors 116 that provide information about the current configuration and / or force at that joint. In the case that the joint is a rotary joint, one or more of the sensors 116 may be torque sensors. In the case that the joint is configured to provide linear motion, one or more of the sensors 116 may be strain gauges. In addition to configuration and / or force data, one or more sensors 116 may additionally provide information about temperature, current, or pressure (such as hydraulic pressure). In one example, one or more configuration sensors are position sensors. That is, the configuration sensors can measure the physical position of the joint. In this example, one or more of the sensors 116 may be encoders. An encoder is a position sensor that converts position measurements into electronic signals. The encoder 116 may be a linear or rotary encoder.

[0033] The drive sources 114 are positioned proximal to the joints that drive their movement in order to improve weight distribution. For clarity, Figure 1Only some of the drive sources and sensors are shown. It can also be understood that the surgical robot 100 may include more than [other components]. Figure 1 Showing more or fewer limbs and joints.

[0034] The surgical robot 100 includes an instrument 112 comprising an end effector for performing surgery. The end effector can take any suitable form. For example, it can be a smooth jaw, a serrated jaw, a gripper, a pair of scissors, a suture needle, a camera, a laser, a scalpel, a stapler, a cauterizer, or a suction device. The instrument also includes an instrument axis and a hinge located between the instrument axis and the end effector. The hinge includes several joints that allow the end effector to move relative to the instrument axis.

[0035] Configuration controllers for the drive source 114 and sensors 116 are distributed within the robot arm 100. The controllers are connected to a control unit 118 via a communication bus. The control unit 118 includes a processor 120 and a memory 122. The memory 122 stores software in a non-transitory manner, which can be executed by the processor 120 to control the operation of the drive source 114 to operate the arm 100. Specifically, the software can control the processor 120 to drive the drive source (e.g., via a distributed controller) based on inputs from sensors 116 and from the surgeon command interface 124. The control unit 118 is coupled to the drive source 114 to drive them based on the outputs generated by the execution of the software. The control unit 118 is electrically connected to the sensors 116 to receive sensed inputs from the sensors and is electrically connected to the command interface 124 to receive inputs from it. (Reference) Figure 1 The electrical connections described in all subsequent figures may be provided, for example, by cables or fiber optic cables, or via wireless connections.

[0036] Command interface 124 includes one or more input devices, thereby allowing a user to request movement of the end effector in a desired manner. The input device or input controller may be, for example, a manually operated mechanical input device, such as a control handle or joystick; a touch-operated input device, such as a touchscreen; or a contactless input device, such as an optical gesture sensor or voice sensor. The input device may monitor eye movement to receive input. The input device may be, for example, some combination of these types of input devices. Commands input by the input device may include movement commands, such as to move the instrument in a specific manner, such as lateral movement and / or rotation. Such commands may include end effector commands, such as to control an end effector coupled to the distal end of instrument 112 to operate the end effector (such as opening / closing a clamp) or to operate (open or close) an electrosurgical end effector.

[0037] The software stored in memory 122 is configured to respond to these inputs according to a predetermined control strategy, and to move the joints of the arm and the instrument accordingly. The control strategy may include safety features that regulate the movement of the arm and instrument in response to command inputs. Therefore, in general, a surgeon at a surgeon's console that appropriately includes a command interface 124 can control the movement of the instrument 112 in a manner that performs the desired surgical procedure. Thus, the robotic arm 100 functions as a master-slave manipulator, with the control unit 118 acting as the master controller. The control unit 118 and / or the command interface 124 may be located remotely from the arm 100.

[0038] Figure 2 This illustrates the use of joints (such as) in a robotic arm for controlling the joints. Figure 1 The system is configured with joints 106a-e shown. The system includes a joint controller 202, a drive source 204, a transmission system 206, and sensors 208, 210, and 212.

[0039] A joint controller 202 is provided to control the configuration of the joint. The joint controller 202 includes one or more electrical connections for connection to a drive source 204 associated with the joint and sensors 208, 210, 212. The joint controller 202 is electrically connected to the sensors 208, 210, 212 via one or more feedback loops 216. The feedback loops 216 enable the controller 202 to receive measurements from the sensors indicating the joint configuration. The joint controller 202 is connected to the drive source 204 via a feedforward loop 214. The joint controller 202 is also connected to a control unit 118 via the feedforward loop and is configured to receive a reference configuration value q from the control unit 118. ref In one example, joint controller 202 is a separate entity from control unit 118. In another example, joint controller 202 and control unit 118 are included in the same entity. Therefore, in this example, control unit 118 can be directly connected to sensors 208, 210, and 212. Joint controller 202 can be distributed within the robot arm such that it is close to the joint it is configured to control. In one example, joint controller 202 is configured to control a single joint. In another example, joint controller 202 is configured to control more than one joint.

[0040] The joint controller 202 is configured to use a reference position value q received from the control unit 118. ref The calculation is performed using the feedback value from sensor 208 to generate a reference torque value τ. ref Reference torque value τ ref Instructions to move the joint to the reference configuration q ref The required torque. In the example where the drive source 204 is a motor, the reference torque value τ is... refThis is used to calculate the desired current value to be supplied to the drive source 204. This desired current is used to generate the output torque τ. o This output torque is the actual torque experienced by the joint when it is driven by the drive source 204. The drive source 204 can correspond to... Figure 1 The driver source 114 is shown.

[0041] A drive source 204 is connected at a first end to a joint controller 202 and at a second end to a drivetrain 206. The drivetrain is configured to transmit power from the drive source 204 to the joint's output. In one example, the drivetrain includes one or more gears. The one or more gears are used to increase or decrease the torque experienced by the joint from the torque supplied by the drive source 204. In a more specific example, the drivetrain is a harmonic drive.

[0042] The system also includes multiple sensors 208, 210, and 212. These multiple sensors 208, 210, and 212 can correspond to... Figure 1 One or more sensors 116 are shown. The plurality of sensors include a first sensor 208, a second sensor 210, and a third sensor 212.

[0043] The first sensor 208 is configured to generate a first type of measurement. The first type of measurement indicates the input configuration q of the joint. i Enter configuration q i Measurements are taken at the output of the drive source 204, which is configured to drive the joint. Therefore, the first sensor 208 can be located at or near the first end of the transmission system 206 where the drive source 204 is located. The first sensor is electrically connected to the joint controller 202 via feedback loop 216. Input configuration q i It can be the input position of the joint.

[0044] The second sensor 210 is configured to generate a second type of measurement. The second type of measurement is represented at joint q. o Output configuration at the input of the joint. Measure the output configuration q at the input of the joint. o Therefore, the second sensor is located at or near the second end of the drive system 206 where the joint is located. The second sensor 210 is electrically connected to the joint controller 202 via a feedback loop 216. Output configuration q o It can be the output position of the joint.

[0045] The third sensor 212 is configured to generate a third type of measurement. The third type of measurement indicates the output torque τ of the joint. o The output torque τ is measured at the input end of the joint. oTherefore, the third sensor is also located at or near the drive system 206. The third sensor 212 is electrically connected to the joint controller 202 via a feedback loop 216.

[0046] Sensors 208, 210, and 212 can be configured to continuously sense data indicative of joint performance. This continuous sensing of data can be performed at regular time intervals. That is, the sensing of data from the joint can be performed by the sensors at a predetermined sampling rate. This sampling rate can be, for example, at... Figure 1 The command interface 124 shown is configurable. With the input and output configurations representing the input and output positions of the joint, respectively, the first sensor 208 and the second sensor 210 can be position sensors, such as linear or rotary encoders.

[0047] In theoretical modeling, the transmission system of an actuated robot joint can be modeled as "rigid." A rigid transmission system is one that does not experience any static friction or recoil. Recoil refers to the idle time experienced due to changes in joint orientation and is caused by gaps or clearances between interface components within the transmission system. If there is no static friction or recoil in the transmission system, then the torque generated by the drive source (i.e., the input torque provided at the first end of the transmission system) will always be proportional to the torque experienced by the joint (i.e., the output torque at the second end of the transmission system). The method used to determine joint stiffness is to correlate the difference between the input and output configurations with the output torque τ measured for the joint. o Comparisons are made. Ideally, the joint should have high stiffness and minimal difference between the input and output configurations.

[0048] In reality, the components of the transmission system that actuates a robot's joints are non-rigid and possess an associated level of elasticity. When a non-rigid joint is driven, it experiences the output torque τ provided by the drive source. o Conversely, inertia. Inertia can be caused by many factors, both inside and outside the joint. An exemplary cause of inertia is the weight of the joint due to gravity. Another exemplary cause is the weight of the object supported by the joint or connected to the joint interface. Due to the presence of inertia, the configuration q of the joint at its output end... o It can be different from the reference configuration q ref For a resilient joint, there exists a degree of resilientness associated with both the drive system 206 and the torque sensor 212. Parameters and disturbances associated with this degree of resilientness must be considered, and these parameters and disturbances are highly variable between different joints. These variations make it difficult to generate a uniform control scheme that can handle joint differences and identify the causes of disturbances.

[0049] As mentioned above, the mechanisms used to drive joints within a robotic system include associated elastic elements, resulting in nonlinear stiffness characteristics. In particular, this elasticity is a characteristic of harmonic drives. Figure 3 This is a graph illustrating an exemplary nonlinear stiffness characteristic of an elastic rotary joint in a robotic arm. The graph plots the elongation ε on the x-axis and the output torque τ on the y-axis. o . τ o The output torque can be measured by a sensor located at the second end of the drivetrain joint. The sensor used to measure the output torque can be a gauge sensor. The gauge sensor is configured to measure the voltage change due to mechanical strain. The elongation is calculated as the input configuration q of the joint. i With output configuration q o The difference between them. In other words, elongation can be defined by the following equation:

[0050] ε=q i -q o

[0051] Where q i Measured at the output of the driver source, while q o Measured at the input end of the joint. In the example below, the elongation is calculated using the equation above. However, it should be understood that it can also be calculated by subtracting the input configuration from the output configuration (i.e., ε = q). o -q i The elongation rate is defined by ().

[0052] exist Figure 3 In the middle, the output torque τ o The elongation is measured in Nm, while the elongation ε is measured in degrees. In an alternative example, the elongation ε can be measured in radians. In both examples, the first and second sensors used to measure the input and output configuration are rotary position sensors. In another example, the elongation can be a linear measurement, which can be calculated using measurements obtained from a linear position sensor. In this example, the elongation can be measured in meters. In yet another example, the elongation can be dimensionless. An example of a dimensionless measurement of elongation is a percentage.

[0053] Figure 3 The graph shown includes five different regions 302 to 310. Regions 302 and 310 illustrate the relationship between the elongation of the transmission system and the output torque τ. oThe direct proportional relationship between them. That is, as the elongation of the transmission system increases, the output torque increases at a constant rate. Region 302 shows the joint performance for negative elongation values, while region 310 shows the joint performance for positive elongation values. That is, region 302 shows the joint performance when the joint moves in a first direction, while region 310 shows the performance when the joint moves in a second direction opposite to the first direction. The gradient of the data graph in regions 302 and 310 represents the stiffness K of the transmission system within the associated elongation range. Therefore, the output torque τ o Elongation (ε = q) i -q o The relationship between the stiffness K and the stiffness K can be characterized by the following equation:

[0054] τ o =K(q) i -q o )

[0055] Therefore, the characteristic of the output torque lies in the elongation (q) of the transmission system. i -q o The direct proportional relationship between () and stiffness K.

[0056] Region 306 represents the recoil phenomenon, where the gradient of the data plot representing the stiffness of the transmission system is close to zero. As mentioned above, recoil refers to the freewheeling experienced by the mechanism when changing direction, and is caused by gaps or clearances between the interface components of the transmission system. For harmonic drives, recoil is a result of the gap between its circular spline and its flexible spline and / or the elastic deformation of the flexible spline under torque. This phenomenon typically occurs in the region of elongation. Figure 3 In the example shown, the elongation value falls between -0.1 degrees and 0 degrees. It should be understood that for different joints, the recoil region will exhibit its own characteristics within different elongation ranges. In this region of the graph, the measured sensor torque is close to zero and does not change as the elongation increases between -0.1 and 0 degrees. Therefore, during the reversal of the joint direction between elongation rates of -0.1 and 0 degrees, no torque is transmitted from the drive source to the joint via the transmission system. Recoil also occurs in the opposite drive direction when the elongation decreases from 0 to -0.1 degrees.

[0057] Regions 304 and 308 are transition regions, or low displacement stiffness regions. Within these regions, the transmission system transitions between recoil region 306 and regions 302 and 310, exhibiting changes in transmission system elongation and output torque τ. oThere is a direct proportional relationship between them. During these regions, there is a non-linear relationship between elongation and sensor torque. These regions occur in joint configurations where not all components of the drive system are in contact with each other, and only a small fraction of the torque from the drive source is transmitted to the joint.

[0058] exist Figure 3 The representative stiffness of the transmission system across the entire elongation range recorded in the graph is represented by the gradient of the regression line 312 passing through all regions of the graph. As described above, and in Figure 3 The study clarifies that the elastic joint exhibits a nonlinear relationship between elongation and output torque. That is, the gradient of stiffness characteristics in the recoil region 306 (which is close to zero) differs substantially from its gradients in regions 302 and 310. Therefore, the straight line represented by 312 does indeed provide an accurate representation of the stiffness characteristics of the transmission system for all elongation values. Furthermore, the constantly changing relationship between elongation and output torque makes it difficult to model and predict the joint's performance throughout its entire range of motion.

[0059] The goal is to provide an accurate indication of when a given joint operates within or below a performance threshold by measuring the performance of its drive system. Therefore, it is important to design a standardized method to characterize joints in this way and to determine whether these joints (and the robots containing them) are suitable for use.

[0060] Figure 4 A method for characterizing the performance of joints in a surgical robotic system is shown. Figure 4 The method shown can also be described as a method for testing joint performance. The method begins at step 402, where a first command signal is sent to the robot arm. This first command signal can be sent to the joint controller 202. The first command signal includes instructions to position the robot arm to an initial configuration. The first command signal may include a first reference configuration q. ref First configuration q ref This can be calculated by applying one or more inverse kinematics calculations to the end effector pose of the command. In one instance, refer to configuration q. ref This refers to the desired physical position of the joint. The initial configuration of the arm can be a predetermined configuration. A predetermined configuration can be selected from a set of predetermined configurations. This set of predetermined arm configurations can be selected to ensure that the joint configuration achieves the desired torque and elongation values.

[0061] A joint's response to a predetermined force is highly dependent on its configuration at the time the force is applied. That is, the joint's inertial mass will vary with its configuration. In some joint configurations, a small inertial mass may cause the joint to rotate without reaching the desired torque and elongation values. Therefore, it is important to understand the initial joint configuration before the testing procedure begins. The initial configuration can be selected from a range of configurations that will determine the torque and elongation values ​​of interest. Positioning the joint in the predetermined configuration before testing also allows for the standardization and repeatability of the test.

[0062] In response to the first command signal 402, the joint is moved to an initial configuration. In one instance, the initial configuration is a predetermined configuration. The joint movement can be driven by a drive source 204 and a transmission system 206. In one instance, feedback is sent to indicate that the joint is in the initial configuration. The feedback may include measurements obtained from one or more of sensors 208, 210, and 212. Alternatively, it can be determined that the joint is in the initial configuration after a known time period has elapsed. In a steady state, the joint configuration does not change over time.

[0063] Once the joint is determined to be in its initial configuration, a second command signal is sent at step 404. The second command signal can be sent to the joint controller 202. The second command signal includes an instruction to apply a force to the joint. The force applied to the joint causes the joint to displace from its stable state. In one example, the second command signal 404 indicates the desired configuration of the joint. That is, the second command signal 404 may include a second reference configuration q. ref Second reference configuration q ref This can be calculated by applying one or more inverse kinematics calculations to the end effector pose of the command. In one instance, refer to configuration q. ref This refers to the desired physical position of the joint. In one instance, displacement of the joint from its steady state results in vibration of the joint. This vibration may be a result of the natural motion as the joint returns to its steady state. Alternatively, the vibration of the joint can be amplified by the joint controller 202. The vibration of the joint provides a wide range of torque and configuration measurements. In another instance, displacement of the joint does not result in vibration, and the joint can return directly to its steady state after displacement. This direct motion may occur in overdamped joints.

[0064] At step 406, the first time interval T is established. N=T1. The first time interval can be defined at the start time of a clock. The clock can be used to count a predetermined period of time from which sensor measurements are received from the joint. The time interval can begin from the time when the sensor measurement indicates that the joint has moved from a steady state. Alternatively, the time interval can begin when a first command signal is sent, or when a second command signal is sent. The time interval can end when the sensor measurement indicates that the joint has returned to a steady state. The time interval can be predetermined and can be calculated before testing. Alternatively, the user can select the time interval based on observations of when the joint returns to a steady state. The time interval can be started manually by the user. Recording begins after the first time interval T is established. N = The number of time intervals that have elapsed since T1.

[0065] At step 408, a first measurement is received for the first time interval T1. The first measurement indicates the configuration of the drive source at the first end of the transmission system. That is, the first measurement indicates the input configuration q of the joint. i The first measurement is as follows: Figure 2 The first sensor 208 shown measures the position of the joint at its input end. In one example, the input configuration is the position of the joint at its input end.

[0066] At step 410, a second measurement is received during the first time interval T1. This second measurement indicates the configuration of the joint at the second end of the transmission system. That is, the second measurement indicates the output configuration q of the joint. o The second measurement was performed by... Figure 2 The second sensor 212 measures the position of the joint at its output end. In one example, the output configuration is the position of the joint at its output end.

[0067] At step 412, a third measurement is received for the first time interval T1. The third measurement indicates the torque experienced by the joint at the second end of the transmission system. That is, the third measurement indicates the output torque τ of the joint. o The third measurement was conducted by... Figure 2 The third sensor 210 is shown for measurement.

[0068] At step 414, the elongation value is calculated for the first time interval T1. The elongation value is calculated using the first and second measurements received for the first time interval T1. As described above, the elongation rate is defined by the following equation:

[0069] ε=q i -q o

[0070] Elongation is calculated by subtracting the second measurement from the first measurement.

[0071] At step 416, the output torque τ of the first time interval is compared. oAnd the value of elongation ε. This comparison is used to derive the output torque τ. o The relationship between the output torque τ and the elongation ε. o The relationship between the joint and elongation ε can indicate the stiffness of the joint at the measured elongation value.

[0072] At step 418, it is determined whether the first time interval T1 represents the end of a time period. Figure 4 In the example shown, the time period ends by a predefined final time interval T. final To identify the end of a time period. In an alternative instance, the end of the time period can be identified by a predetermined time (e.g., a predetermined number of seconds after the start of the time period). In another instance, the time period can end automatically when it is determined that the joint has returned to a steady state. That is, when the sensor data no longer indicates a change in joint configuration over time. In yet another instance, the time period may not be predefined and can be manually terminated by the operator.

[0073] If it is determined that the first time interval T1 does not represent the end of the time period, the method proceeds to step 420. At step 420, a new time interval is selected. That is, T... N Replace T N+1 In T N If the first time interval T1 is used, then at step 420, T1 is replaced by the second time interval T2. The procedure of receiving the first, second, and third measurements is then repeated within the second time interval T2. The elongation rate for the second time interval is calculated, and then compared with the output torque. It is then determined whether T2 represents the end of the time interval. If T2 does not represent the end of the time interval, the third time interval T3 is selected. Therefore, method steps 408-420 are executed iteratively until a time interval T is determined. N This definitely indicates the end of the time period. If this is the case, the method proceeds to step 422, where output is generated.

[0074] exist Figure 4 In the example shown, step 418 occurs after step 416. In an alternative example, step 418 can occur before step 416. That is, output torque and joint configuration measurements can be extracted, but can only be compared after the time period has ended. Alternatively, step 418 can occur simultaneously with step 416.

[0075] Therefore, the elongation ε and the output torque τ o The value is in each time interval T1, T2...T finalThe comparisons between these two values ​​at each time interval can be combined to derive the relationship between the two parameters. That is, the sensor data obtained at each time interval can be combined to determine the change in output torque as the elongation changes. This relationship can be derived by selecting a time interval of interest within a predetermined time period. This time interval can be selected by identifying a subset of time intervals of interest that represent the range of measured elongation and / or output torque values. The relationship can also be derived by reordering the elongation values ​​such that they are arranged in ascending order of their magnitude, rather than in chronological order. If the elongation values ​​are reordered in this way, then the output torque values ​​associated with each reordered elongation value (i.e., torque measurements received at the same time intervals used to calculate the elongation) are reordered accordingly.

[0076] Elongation ε and output torque τ o The comparison between the measurements can further include comparing the relationship between these two parameters with predetermined characteristics. These predetermined characteristics can be obtained from predetermined theoretical or experimental data. Predetermined characteristics can indicate the desired performance of the joint. Therefore, if the elongation ε is related to the output torque τ... o If the derived relationship between the joints matches the predetermined characteristics within a predetermined threshold, then the joint can be determined to perform within its expected performance threshold. Therefore, the joint can be determined to be suitable for use. If the joint's performance does not match the predetermined characteristics within the predetermined threshold, then the joint can be determined to perform below its expected performance threshold.

[0077] At step 422, an output indicating the joint's performance is generated. This output indicates whether the joint is performing within its desired performance threshold. In one instance, the output provides a yes / no indication of whether the joint is performing within the desired performance threshold. In this first instance, the output is audible. Alternatively or additionally, the output may be visual. The visual output may be an LED indicator or an alternative indicator light. The visual output may be provided on a display. The output may provide a detailed analysis of the joint's performance. The output may include commands to update the joint's control parameters.

[0078] The second command signal 404 to be provided to the robot arm can be referred to as the excitation function. Examples of different excitation functions that can be used to shift a joint from a steady state are provided below:

[0079] 1. The second command signal may include instructions to apply an instantaneous force to the joint and displace the joint from its steady state. This type of function is also called a step function. A first, second, and third measurement can then be received at multiple predefined time intervals as the joint returns to its steady state. This time interval should end when the joint returns to its steady state. The instantaneous force can have a large amplitude; that is, the instantaneous force can constitute a violent motion with high acceleration. This allows the joint to experience the full range of elongation and torque values ​​as it returns to its steady state. This function is preferably used on joints located closer to the base unit 102 of the robot arm. Joints closer to the base unit are associated with higher inertia values.

[0080] 2. The second command signal may include an instruction to continuously increase the frequency of the force applied to the joint over time. This continuous increase in frequency can occur within a predetermined time period. This type of function is also known as a linear frequency modulation function. The first, second, and third measurements can then be received at multiple predefined time intervals. This excitation function produces a large output torque value for the joint, but also generates a large amount of data over a wide frequency range. This function is preferably used for joints located away from the base unit 102 of the robot arm or close to the mechanism 112. Joints away from the base unit 102 experience higher inertia.

[0081] 3. The second command signal may include an instruction to consistently change the direction of the force to be applied to the joint over a predetermined time period. This continuous change in direction will result in acceleration, which will alter the force applied to the joint. A first measurement, a second measurement, and a third measurement can then be received at multiple predefined time intervals. As in Example 2 above, this function is preferably used for joints located away from the base unit 102 of the robot arm or close to the mechanism 112.

[0082] As mentioned above, the choice of excitation function for testing depends largely on the joint being observed. Joints closer to the base unit 102 of the robot arm experience higher inertia and therefore benefit from testing with high-amplitude forces. Joints closer to the mechanism 112 experience lower inertia and therefore benefit from testing with gradually increasing forces.

[0083] In one example, generating output indicating joint performance includes calculating an elongation versus torque curve to define the joint's stiffness characteristics. This curve can be plotted and visualized on a display. The plotted curve can correspond to... Figure 3 The graph shown.

[0084] This graph can be output to a display located outside the robotic arm. The graph allows observers to view, compare, and analyze the results obtained from the test. Therefore, the graph can provide a more detailed indication of the joint's performance. In one example, the graph calculating elongation versus torque also includes one or more regression lines on the graph, each associated with a corresponding gradient. The regression line is the line that best fits its corresponding dataset. As mentioned above, the gradient of the output torque versus elongation data graph represents the stiffness of the transmission system for the range of elongation values ​​within that data graph. Figure 3 In Figure 3 The representative stiffness of the transmission system is represented by the gradient of regression line 312 across the entire range of elongation values ​​recorded. Figure 3 As can be seen, for an elastic joint, the relationship between elongation and output torque is not entirely linear. That is, the gradient of the stiffness characteristic in the recoil region 306 (where it is zero) is largely different from its gradient in regions 302 and 310. Therefore, the straight line represented by 312 does indeed provide an accurate representation of the stiffness characteristics of the transmission system for all elongation values.

[0085] Therefore, the more accurate stiffness characteristics of the joint's transmission system can be determined by deriving multiple regression lines from the elongation versus torque curve. Multiple gradients can be used to identify multiple different regions on the curve. Each region is associated with a different stiffness value. The regions can be quantitatively described by curve / line fitting and compared with some optimal / expected thresholds and ideal curves. Alternatively, regions can be identified by applying signal conditioning to the received sensor measurements. Signal conditioning enables region detection with greater accuracy.

[0086] In one example, the elongation versus torque curve can be divided into three regions. These three regions can be defined as follows:

[0087] 1. A first linear relationship between elongation and torque in a first direction;

[0088] 2. A second linear relationship between elongation and torque in a second direction; and

[0089] 3. Recoil zone.

[0090] Predefined "breakpoints" can be used to identify the first, second, and third regions of a graph. These "breakpoints" can be characterized by predetermined elongation or torque values. For example, a first torque value of zero (or close to zero) can identify the beginning of the recoil region in the graph. Alternatively, the first, second, and third regions of the graph can be identified by applying a piecewise fit to the graphical data. That is, new regions can be defined when a significant change in the relationship between output torque and elongation is identified. In this example, the first and second regions, as summarized above, can be identified by an elongation-torque relationship that can be approximated by a regression line with a gradient above a predetermined threshold. In contrast, the recoil region can be identified by a regression line with a gradient below a predetermined threshold. The gradient of the regression line near the recoil region may be close to zero. A gradient close to zero is a gradient in which the torque experiences minimal change with respect to the change in elongation outside the recoil region, even when the change in elongation is large. The graphical data can be characterized by a piecewise function.

[0091] To demonstrate the difference between the gradient of the regression line characterizing the recoil region and the gradient of the regression line outside this region, consider the following example. For a specific joint in a robotic arm, the elongation of its transmission system may vary between 2 degrees and -2 degrees. Within this range, the elongation value of interest, to characterize the joint's performance, can be between -0.5 degrees and 0.5 degrees. The recoil region may appear between -0.1 degrees and 0.1 degrees. Therefore, relative to the total range of elongation values ​​of interest, the recoil region appears in those larger subranges of values. That is, the range of elongation values ​​in the recoil region is larger relative to the entire range of elongation values ​​of interest. In this example, the range of torque values ​​of interest for a specific joint can vary from -15 Nm to 15 Nm. The range of torque associated with the recoil region can vary between -0.1 Nm and 0.1 Nm. That is, the torque variation in the recoil region is small relative to the total observed range of torque values ​​of interest. Therefore, on the elongation versus torque graph, the range of elongation values ​​is relatively large in the recoil region, while the range of torque values ​​is relatively small. The gradient of the regression line characterizing torque as a function of elongation in the recoil region can be close to zero relative to the gradient of the regression line outside this region (e.g., in the first and second regions described above). For example, for a specific joint in a robotic arm, the gradient of the regression line characterizing the recoil region can have a maximum value of 15 Nm / degree. In other words, the recoil region can be identified by a regression line with a gradient below a predetermined threshold of 15 Nm / degree. Normal stiffness regions (e.g., the first and second regions described above) can be identified by regression lines with a gradient above this predetermined threshold of 15 Nm / degree. More commonly, normal stiffness regions can have a gradient greater than 50 Nm / degree. Therefore, the gradient of the regression line in the recoil region can be described as close to zero relative to the gradient of its surrounding region. The above values ​​are provided only as examples, as it can be understood that for different joints, the recoil region will present itself within different ranges of elongation values.

[0092] In alternative examples, the elongation versus torque curve can be divided into five regions. These five regions can be defined as follows:

[0093] 1. A first linear relationship between elongation and torque in a first direction;

[0094] 2. In the first low displacement stiffness region in the first direction;

[0095] 3. A second linear relationship between elongation and torque in the second direction;

[0096] 4. In the second low displacement stiffness region in the second direction; and

[0097] 5. Recoil zone.

[0098] Reference Figure 3 These five regions are described in more detail. As mentioned above, the five distinct regions can be identified using "breakpoints" or piecewise fitting. It is understood that more or fewer regions than those specified herein can be derived from graphs indicating the stiffness characteristics of a joint. For example, some joints may have more than one associated recoil region within a range of elongation values.

[0099] The graph can be further analyzed to determine the accuracy of one or more regression lines. In one instance, R-squared statistics can be used to determine the accuracy of one or more regression lines. R-squared statistics are used to assess the distribution of data points around the regression line and determine the accuracy of the regression line in representing graphical data. Therefore, R-squared analysis can be used to determine the accuracy of one or more regression lines representing graphical data. In alternative instances, signal-to-noise ratio, multinomial, least squares, or any other known analytical method can be used to determine the accuracy of one or more regression lines.

[0100] In one instance, Figure 4 The method described herein can be commanded by a test unit located outside the robotic arm. The test unit may include components capable of performing actions such as... Figure 4 The software provides the test instructions. That is, the test unit can be connected to a robotic arm, enabling it to send first and second command signals to the robotic arm and receive sensor measurements from the robotic arm. Alternatively, Figure 4 The method described herein can be controlled by control unit 118. In this example, the user can instruct the start of the test at command interface 124. Control unit 118 can then send a first command signal and a second command signal to the robot arm and receive sensor measurements from the robot arm as instructed. The control unit may include functions for performing... Figure 4 The method described herein uses several specific control parameters. These control parameters may differ from those used during normal operation of the robotic arm. Utilizing the concepts of this invention, it is possible to obtain a characteristic curve of the joint during the process prior to system operation or during power-on self-test (POST), compare this curve with a curve of an ideal joint, and determine whether the joint is suitable for use.

[0101] Figure 4 The method described herein can be executed during the initialization protocol of the robotic arm. Alternatively, the method can be executed during routine testing of the arm. The method can also be executed during the manufacturing of the arm.

[0102] The method described above can be used to fully characterize joint performance using sensors and drive mechanisms integrated within the robot arm. That is, no additional sensors or equipment are needed to characterize the joint's transmission system. This differs from robot systems that do not include two configured sensors. For such systems, it is necessary to lock one end of the transmission system while forcing the other end to rotate in order to measure the torque at the unlocked end of the transmission system and thus obtain the joint's stiffness characteristics. Therefore, the method described herein provides a simple, fast, and reliable way to characterize the performance of joints in a robot arm. Based on the derivation of joint performance, control parameters can be varied during joint operation to illustrate that performance. Alternatively, this derivation can be used to determine if a joint is unsuitable for use or requires maintenance.

[0103] As mentioned above, Figure 3 and Figure 4 The force applied to and measured from a joint is called torque. Therefore, Figure 3 and Figure 4 Methods for configuring and measuring rotary joints are disclosed. In an alternative embodiment, the joint to be controlled can be configured to provide linear motion. In this embodiment, strain gauges or other force sensors can be used to measure and calculate stress.

[0104] The applicant hereby independently discloses each individual feature described herein, as well as any combination of two or more such features, provided that such features or combinations can be implemented based on this specification as a whole according to common general knowledge of those skilled in the art, regardless of whether such features or combinations of features solve any problem disclosed herein, and without limiting the scope of the claims. The applicant notes that aspects of the invention can consist of any such single feature or combination of features. In view of the foregoing description, it will be apparent to those skilled in the art that various modifications can be made within the scope of the invention.

Claims

1. A method for characterizing the performance of a joint in a surgical robotic arm, said joint being driven by a transmission system that transmits power from a drive source to said joint, said method comprising: Send a first command signal to position the robotic arm in its initial configuration; A second command signal is sent to apply force to the joint, thereby displacing the joint from a stable state; For multiple predefined time intervals: Receive a first measurement indicating the configuration of the drive source at a first end of the transmission system; Receive a second measurement indicating the configuration of the joint at the second end of the transmission system; The elongation value is calculated using the first and second measurements; as well as Receive a third measurement, the third measurement indicating the torque experienced by the joint at the second end of the transmission system; The elongation value is compared with the corresponding torque value at each of the predefined time intervals; as well as The comparison generates an output indicating the performance of the joint.

2. The method of claim 1, wherein generating the output includes calculating a graph of elongation versus torque to define the stiffness characteristics of the joint.

3. The method of claim 1, further comprising comparing the elongation value and the corresponding torque value with predetermined characteristics of the joint.

4. The method of any one of claims 1 to 3, wherein the method comprises: Send a command signal to apply an instantaneous force to displace the joint from the stable state; as well as The first measurement, the second measurement, and the third measurement are received at multiple predefined time intervals when the joint returns to the stable state.

5. The method of any one of claims 1 to 3, wherein the method further comprises: Send command signals to consistently increase the frequency of the force applied to the joint; as well as The first measurement, the second measurement, and the third measurement are received at multiple predefined time intervals.

6. The method of any one of claims 1 to 3, wherein the method further comprises: Send command signals to consistently change the direction of the force to be applied to the joint; as well as The first measurement, the second measurement, and the third measurement are received at multiple predefined time intervals.

7. The method of any one of claims 1 to 3, wherein the second command signal indicates the desired configuration of the joint.

8. The method of any one of claims 1 to 3, wherein elongation is characterized as a difference between a measurement configuration of the drive source and a measurement configuration of the joint. ​ 9. The method of claim 2, further comprising calculating one or more regression lines on the graph of elongation versus torque, each regression line having a corresponding gradient.

10. The method of claim 9, wherein the method further comprises analyzing the gradient of the one or more regression lines to identify multiple distinct regions.

11. The method of claim 10, wherein the method further comprises identifying a recoil region from the graph, the recoil region being characterized by a regression line having a near-zero gradient.

12. The method of claim 9, wherein the method further comprises identifying one or more linear stiffness regions from the graph, each linear stiffness region being identified by a regression line having a gradient above a predetermined threshold.

13. The method of any one of claims 9 to 12, further comprising analyzing the graph to determine the accuracy of the one or more regression lines characterizing the data of the graph.

14. The method of claim 13, wherein analyzing the graph includes performing R-squared analysis.

15. The method of claim 10, wherein the plurality of different regions are identified by applying signal conditioning to the received sensor measurements.

16. The method of any one of claims 1 to 3, wherein the drive source is an electric motor.

17. The method of any one of claims 1 to 3, wherein the initial configuration of the arm is selected from a set of predetermined arm configurations.

18. The method of any one of claims 1 to 3, wherein the transmission system comprises one or more gears.

19. The method of any one of claims 1 to 3, wherein the transmission system is a harmonic drive.

Images