Dexterous hand data acquisition method and device, chip and storage medium
By using time-sharing reading and data processing methods, the problem of low efficiency in acquiring sensor data by dexterous hands is solved, achieving efficient sensor data processing and precise control of dexterous hands, thus adapting to the needs of object grasping in complex environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SUTENG INNOVATION TECHNOLOGY CO LTD
- Filing Date
- 2025-06-13
- Publication Date
- 2026-07-21
AI Technical Summary
How to efficiently acquire the sensor data collected by the position sensor and tactile sensor in the dexterous hand to achieve high-precision motion control and tactile feedback, especially to improve the reliability of object grasping in complex environments.
Using a time-division reading method, the sensor data of the position sensor and tactile sensor in the dexterous hand finger mechanism are read separately in different cycles through the same interface. Combined with timed interrupt signals and data processing, a tactile sensor data packet is generated and sent to the task control system.
It improves the reading rate and real-time performance of sensor data, adapts to the update rate requirements of different types of sensor data, saves communication bandwidth, and enhances the control precision and reliability of the dexterous hand.
Smart Images
Figure CN120791734B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data acquisition, and in particular to methods, devices, chips, and storage media for data acquisition using dexterous hands. Background Technology
[0002] In the field of embodied intelligent robots, a dexterous hand that can respond quickly and control precisely is one of the key elements. Position sensors can monitor the movement of the dexterous hand's finger joints in real time (such as angle and displacement), and working in conjunction with the dexterous hand's drive system and transmission structure, high-precision motion control can be achieved. Meanwhile, tactile sensors can provide multi-dimensional force feedback (pressure, friction, etc.), improving the ability and reliability of object grasping in complex environments (such as when visual perception is interfered with by lighting or when objects have irregular shapes).
[0003] Each finger of a dexterous hand is equipped with a position sensor and a tactile sensor. How to obtain the sensor data collected by these sensors to determine the state of the dexterous hand and control it has become an urgent technical problem to be solved. Summary of the Invention
[0004] This application provides a method, apparatus, chip, and storage medium for acquiring data from a dexterous hand, with the aim of obtaining sensor data within the dexterous hand.
[0005] Firstly, a data acquisition method for a dexterous hand is provided, applied to a dexterous hand, wherein the dexterous hand includes multiple finger mechanisms, each finger mechanism being provided with a first tactile sensor, a second tactile sensor, and multiple position sensors, the first tactile sensor, the second tactile sensor, and the multiple position sensors within the same finger mechanism being connected to the same interface, the method comprising:
[0006] During the first cycle, the sensing data collected by each sensor in the first sensor group is read in time-division through the first interface. The first interface is an interface connected to the sensor in the first finger mechanism. The first finger mechanism is any one of the plurality of finger mechanisms. The first sensor group includes a plurality of position sensors and a first tactile sensor in the first finger mechanism.
[0007] During the second cycle, the sensing data collected by each sensor in the second sensor group is read in a time-division manner through the first interface. The second cycle is the next cycle after the first cycle. The second sensor group includes multiple position sensors and a second tactile sensor in the first finger mechanism.
[0008] In this technical solution, multiple position sensors and multiple tactile sensors within the same finger mechanism of the dexterous hand are connected to the same interface. In the first cycle, the sensor data collected by each sensor in the first sensor group is read in a time-division manner through the first interface. In the second cycle, the sensor data collected by each sensor in the second sensor group is read in a time-division manner through the first interface. The second cycle is the cycle following the first cycle. Since the first sensor group includes multiple position sensors and a first tactile sensor within the first finger mechanism, and the second sensor group includes multiple position sensors and a second tactile sensor within the first finger, the system reads the sensor data collected by all position sensors and one tactile sensor within the finger mechanism in each cycle. This allows for the acquisition of all position sensor data corresponding to the finger mechanism in each cycle and all tactile sensor data corresponding to the finger mechanism in multiple cycles. The reading rate of position sensor data within the finger mechanism can be increased based on the priority of data acquisition, thereby adapting to the update rate requirements of different types of sensor data and saving communication bandwidth.
[0009] In conjunction with the first aspect, in one possible implementation, the step of reading the sensor data collected by each sensor in the first sensor group through the first interface in a time-division manner during the first period includes: in response to a target time interrupt signal, reading the sensor data collected by each sensor in the first sensor group through the first interface in a time-division manner, wherein the target time interrupt signal is a time interrupt signal with the highest priority.
[0010] By reading the sensor data collected by multiple position sensors and one tactile sensor within the finger mechanism based on the highest priority timer interrupt signal, the highest priority of sensor data reading can be guaranteed, which is beneficial to the real-time performance and accuracy of sensor data.
[0011] In conjunction with the first aspect, in one possible implementation, after reading the sensing data collected by each sensor in the first sensor group through the first interface in a time-division manner during the first period, the method further includes: performing data calculation on the sensing data collected by the position sensor in the first sensor group to obtain the position data of the first finger mechanism during the first period.
[0012] The position data of the finger mechanism is obtained by performing data calculations on the position sensor data collected within a cycle, which can improve the real-time performance and accuracy of the finger mechanism's position data.
[0013] In conjunction with the first aspect, in one possible implementation, after reading the sensing data collected by each sensor in the second sensor group through the first interface in a time-division manner during the second cycle, the method further includes: generating a tactile sensing data packet based on the sensing data collected by each tactile sensor in the plurality of finger mechanisms during the first cycle and the second cycle; and sending the tactile sensing data packet to the task control system.
[0014] After acquiring the sensor data collected by the tactile sensors of each finger through multiple cycles, a tactile sensor data packet is generated and sent to the task control system so that the task control system can generate corresponding control tasks or control commands, thereby realizing the control of the dexterous hand.
[0015] In conjunction with the first aspect, in one possible implementation, generating a tactile sensing data package based on the sensing data collected by each tactile sensor in the plurality of finger mechanisms during the first cycle and the second cycle includes: obtaining a first sensing dataset from the sensing data collected by each tactile sensor in the plurality of finger mechanisms during the first cycle and the second cycle, the first sensing dataset including multiple sensing data belonging to the same finger mechanism; and generating the tactile sensing data package based on the interface identifier corresponding to the first sensor dataset, the multiple sensing data in the first sensing dataset, and the sensor identifiers corresponding to each of the multiple sensing data in the first sensing dataset.
[0016] A tactile sensor data package is generated based on the sensor data collected by the tactile sensor of a finger mechanism. While reducing the amount of data in the data package, it makes it easier for the task control system to directly perceive the tactile state of each finger mechanism based on each tactile sensor data package.
[0017] In conjunction with the first aspect, in one possible implementation, before generating the tactile sensing data packet based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, the method further includes: generating a verification code based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset; the step of generating the tactile sensing data packet based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset includes: generating the tactile sensing data packet based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, and the verification code.
[0018] Before generating the data packet, a checksum is generated, and a tactile sensing data packet is generated based on the checksum and the sensing data so that the task control system can verify the accuracy of the sensing data.
[0019] In conjunction with the first aspect, in one possible implementation, the method further includes: during the first period, reading the sensor data collected by each sensor in the third sensor group in a time-division manner through a second interface; wherein the second interface is an interface connected to a sensor in the second finger mechanism, the second finger mechanism is any finger mechanism other than the first finger mechanism among the plurality of finger mechanisms, and the third sensor group includes a plurality of position sensors and a first tactile sensor in the second finger mechanism; performing data processing on the sensor data collected by the position sensors in the third sensor group to obtain the position data of the second finger mechanism during the first period.
[0020] Within the same cycle, it also acquires and processes sensor data collected by multiple position sensors of other finger mechanisms through other interfaces, enabling it to obtain position data of all fingers of the dexterous hand within one cycle, thus facilitating better control of the dexterous hand.
[0021] Secondly, a dexterous hand data acquisition device is provided for use in a dexterous hand. The dexterous hand includes multiple finger mechanisms, each of which is provided with a first tactile sensor, a second tactile sensor, and multiple position sensors. The first tactile sensor, the second tactile sensor, and the multiple position sensors in the same finger mechanism are connected to the same interface.
[0022] The device includes:
[0023] The time-division reading module is used to read the sensor data collected by each sensor in the first sensor group in a time-division manner through the first interface during the first cycle. The first interface is an interface connected to the sensor in the first finger mechanism. The first finger mechanism is any one of the plurality of finger mechanisms. The first sensor group includes a plurality of position sensors and a first tactile sensor in the first finger mechanism.
[0024] The time-division reading module is also used to read the sensing data collected by each sensor in the second sensor group through the first interface in a time-division manner during the second cycle, wherein the second cycle is the next cycle after the first cycle, and the second sensor group includes multiple position sensors and a second tactile sensor in the first finger mechanism.
[0025] Thirdly, a master control chip is provided, including a memory and a processor, the memory being connected to the processor, the processor being configured to execute one or more computer programs stored in the memory, and the processor, when executing the one or more computer programs, causing the master control chip to execute the dexterous hand data acquisition method of the first aspect described above.
[0026] Fourthly, a computer-readable storage medium is provided, which stores a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to perform the dexterous hand data acquisition method of the first aspect.
[0027] This application can achieve the following technical effects: in each cycle, it reads the sensing data collected by all position sensors in the finger mechanism and reads the sensing data collected by one of the tactile sensors in the finger mechanism. In this way, it obtains all position sensing data corresponding to the finger mechanism in each cycle and all tactile sensing data corresponding to the finger mechanism in multiple cycles. It can improve the reading rate of position sensing data in the finger mechanism according to the priority of data acquisition, thereby adapting to the update rate requirements of different types of sensing data and saving communication bandwidth. Attached Figure Description
[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0029] Figure 1 and Figure 2 This is a schematic diagram showing how the sensors are connected within a dexterous hand.
[0030] Figure 3 and Figure 4 A schematic diagram illustrating the connection scheme of the sensors in the dexterous hand provided in the embodiments of this application;
[0031] Figure 5 A flowchart illustrating a dexterous hand data acquisition method provided in an embodiment of this application;
[0032] Figure 6 This is a schematic diagram of the structure of a dexterous hand data acquisition device provided in an embodiment of this application.
[0033] Figure 7 This is a schematic diagram of the structure of a main control chip provided in an embodiment of this application. Detailed Implementation
[0034] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.
[0035] It should be noted that, unless there is a conflict, the various features in the embodiments of this application can be combined with each other, all of which are within the protection scope of this application. Furthermore, although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than the module division in the device or the order in the flowchart. Moreover, the terms "first," "second," and "third" used in this application do not limit the data or execution order, but only distinguish identical or similar items with essentially the same function and effect.
[0036] A dexterous hand is a key component in the field of robotics. Its mechanical structure is based on the anatomy of the human hand, enabling the grasping, manipulation, and interaction of complex objects, giving robots the flexibility and adaptability of a human hand. Each finger of a dexterous hand typically has multiple joints, which can flexibly bend and extend to perform various functional movements of the human hand.
[0037] To perform various functions and movements of the human hand, each finger of a dexterous hand is equipped with multiple position and tactile sensors. In some connection schemes, these sensors are used... Figure 1 or Figure 2 The connection method shown connects the position sensor and tactile sensor inside each finger. Figure 1 The connection method shown requires cascading multiple master controllers, which results in high implementation costs. Figure 2 The connection method shown has complex wiring, making it difficult to simultaneously sense the state of the fingers.
[0038] In order to achieve simultaneous sensing of finger status while reducing costs, this application proposes a sensor connection scheme in a dexterous hand. Based on the sensor connection scheme in a dexterous hand, a data acquisition scheme for a dexterous hand is proposed, aiming to better acquire the sensing data collected by the sensors in the dexterous hand.
[0039] First see Figure 3 and Figure 4 , Figure 3 and Figure 4 A schematic diagram of the sensor connection scheme in the dexterous hand provided in the embodiments of this application is shown below. Figure 3 and Figure 4As shown, the dexterous hand 1 includes multiple finger mechanisms 10 and a main controller 20. The multiple finger mechanisms 10 include a thumb finger mechanism 101, an index finger finger mechanism 102, a middle finger finger mechanism 103, a ring finger finger mechanism 104, and a little finger finger mechanism 105. Each finger mechanism 10 is equipped with multiple position sensors and multiple tactile sensors. The multiple position sensors and multiple tactile sensors within the same finger mechanism are connected to the same interface. The multiple tactile sensors within the same finger mechanism include a first tactile sensor and a second tactile sensor, which are two tactile sensors within the same finger mechanism. For example, each joint of the finger mechanism is equipped with a tactile sensor, and the first tactile sensor and the second tactile sensor are tactile sensors within two adjacent joints of the same finger mechanism.
[0040] Multiple position sensors and multiple tactile sensors within the same finger mechanism are connected to the main controller 20 via a single interface. The main controller 20 receives sensor data collected by the sensors within the multiple finger mechanisms 10, and is responsible for the operation of each component, processing sensory information, and executing control commands. The main controller 20 includes, but is not limited to, a microcontroller unit (MCU), a central processing unit (CPU), or a field-programmable gate array (FPGA).
[0041] The interface connecting the main controller 20 with multiple position sensors and multiple tactile sensors in the same finger mechanism can be a bus interface, including but not limited to serial peripheral interface (SPI), I2C interface, etc.
[0042] In some specific embodiments, multiple position sensors and multiple tactile sensors within each finger mechanism, such as Figure 3 As shown, the number of position sensors and tactile sensors within the finger mechanism is the same as the number of knuckle mechanisms in the finger mechanism; each knuckle mechanism is equipped with one position sensor and one tactile sensor. Specifically, the thumb finger mechanism 101 includes two position sensors and two tactile sensors, while the index finger mechanism 102, middle finger mechanism 103, ring finger mechanism 104, and little finger mechanism 105 each include three position sensors and three tactile sensors.
[0043] In some embodiments, such as Figure 3 As shown, the dexterous hand 1 also includes a palm tactile sensor, a back of hand tactile sensor, a thumb swing position sensor, and a thumb pitch position sensor.
[0044] exist Figure 3 and Figure 4 In the connection scheme, since all the sensors in a finger mechanism are connected to the main controller through a single interface, the connection harness can be concentrated in a single finger mechanism, the wiring method is simple, and the main controller can easily sense the state of each finger joint through a single interface.
[0045] based on Figure 3 and Figure 4 The connection scheme enables the data acquisition scheme of the dexterous hand described in this application. This is described in detail below. It is understood that a finger mechanism contains multiple tactile sensors. The following description focuses on two tactile sensors within the finger mechanism, namely the aforementioned first tactile sensor and second tactile sensor. For the definitions of the first and second tactile sensors, please refer to the foregoing description.
[0046] See Figure 5 , Figure 5 This is a flowchart illustrating a data acquisition method for dexterous hands provided in an embodiment of this application. This method is applied to dexterous hands. For an introduction to dexterous hands, please refer to... Figure 3 and Figure 4 Corresponding description; such as Figure 5 As shown, the method includes:
[0047] S301, during the first cycle, reads the sensing data collected by each sensor in the first sensor group through the first interface in a time-division manner.
[0048] The first interface is an interface for connecting to a sensor within the first finger mechanism, which is any one of multiple finger mechanisms in a dexterous hand. Figure 3 Taking the dexterous hand shown as an example, the first finger mechanism can be a thumb finger mechanism, an index finger mechanism, a middle finger finger mechanism, a ring finger finger mechanism, or a little finger finger mechanism.
[0049] The first sensor group includes multiple position sensors and a first tactile sensor within the first finger mechanism. Taking an example where the first finger mechanism includes three position sensors and three tactile sensors, assuming the three position sensors in the first finger mechanism are position sensor 11, position sensor 12, and position sensor 13 in sequence, and the three tactile sensors in the first finger mechanism are tactile sensor 11, tactile sensor 12, and tactile sensor 13 in sequence, then the first sensor group includes position sensor 11, position sensor 12, position sensor 13, and tactile sensor 11; or, the first sensor group includes position sensor 11, position sensor 12, position sensor 13, and tactile sensor 12; or, the first sensor group includes position sensor 11, position sensor 12, position sensor 13, and tactile sensor 13.
[0050] Within the first cycle, reading the sensor data collected by each sensor in the first sensor group through the first interface in a time-division manner means dividing the first cycle into multiple time slices, reading the sensor data collected by one sensor in the first sensor group in one time slice, thereby reading the sensor data collected by all sensors in the first sensor group.
[0051] Taking the first sensor group, which includes the aforementioned position sensor 11, position sensor 12, position sensor 13, and tactile sensor 11, as an example, the first cycle can be divided into at least four time segments. Assuming that the at least four time segments include time segment 1 to time segment 4, then in time segment 1, the sensing data collected by position sensor 11 is read; in time segment 2, the sensing data collected by position sensor 12 is read; in time segment 3, the sensing data collected by position sensor 13 is read; and in time segment 4, the sensing data collected by tactile sensor 11 is read. In this way, the sensing data collected by all sensors in the first sensor group is read.
[0052] In some feasible implementations, in response to a target timer interrupt signal, the sensing data collected by each sensor in the first sensor group can be read in a time-division manner through the first interface, where the target timer interrupt signal is the highest priority timer interrupt signal. For example, the target timer interrupt signal is an interrupt signal triggered by a hardware timer.
[0053] Specifically, it is determined whether a target timer interrupt signal has been received. If a target timer interrupt signal is received, it indicates that the start time of the first cycle has been reached. The system then begins time-division reading of sensor data collected by each sensor in the first sensor group through the first interface, until all sensor data collected by all sensors in the first sensor group has been read. This reading can be done via direct memory access (DMA).
[0054] By reading the sensor data collected by multiple position sensors and one tactile sensor within the finger mechanism based on the highest priority timer interrupt signal, the highest priority of sensor data reading can be guaranteed, which is beneficial to the real-time performance and accuracy of sensor data.
[0055] In some embodiments, sensing data collected by each sensor in the first sensor group can also be read in a time-division manner through the first interface based on other types of timing signals. For example, the start time of the first cycle can also be determined based on a software timing signal, thereby reading the sensing data collected by each sensor in the first sensor group. This application does not impose any limitations on this.
[0056] S302, during the second cycle, reads the sensing data collected by each sensor in the second sensor group through the first interface in a time-division manner.
[0057] Here, the second cycle is the cycle following the first cycle, meaning the second cycle and the first cycle are two adjacent cycles.
[0058] The second sensor group includes multiple position sensors and a second tactile sensor within the first finger mechanism. For example, if the first sensor group includes the aforementioned position sensor 11, position sensor 12, position sensor 13, and tactile sensor 11, then the second sensor group includes the aforementioned position sensor 11, position sensor 12, position sensor 13, and tactile sensor 12. Similarly, if the first sensor group includes the aforementioned position sensor 11, position sensor 12, position sensor 13, and tactile sensor 13, then the second sensor group includes the aforementioned position sensor 11, position sensor 12, position sensor 13, and tactile sensor 11.
[0059] In the second cycle, the implementation principle of reading the sensor data collected by each sensor in the second sensor group through the first interface in a time-division manner is the same as the implementation principle of reading the sensor data collected by each sensor in the first sensor group through the first interface in a time-division manner in the first cycle. Please refer to the relevant description of step S301 above, which will not be repeated here.
[0060] Understandably, the above Figure 5 The corresponding technical solution is introduced using two tactile sensors (i.e., the first tactile sensor and the second tactile sensor) in the finger mechanism as an example. When there are more than two tactile sensors in the finger mechanism, the sensor data collected by all position sensors and the sensor data of one tactile sensor in each of the multiple cycles can be read in each of the multiple cycles, so as to obtain the tactile sensor data corresponding to the finger mechanism in multiple cycles.
[0061] In the above Figure 5In the corresponding technical solution, during the first cycle, the sensing data collected by each sensor in the first sensor group is read in a time-division manner through the first interface, and during the second cycle, the sensing data collected by each sensor in the second sensor group is read in a time-division manner through the first interface. The second cycle is the cycle following the first cycle. Since the first sensor group includes multiple position sensors and a first tactile sensor in the first finger mechanism, and the second sensor group includes multiple position sensors and a second tactile sensor in the first finger, the sensing data collected by all position sensors in the finger mechanism and one tactile sensor in the finger mechanism are read in each cycle. In this way, all position sensing data corresponding to the finger mechanism are obtained in each cycle, and all tactile sensing data corresponding to the finger mechanism are obtained in multiple cycles. The reading rate of position sensing data in the finger mechanism can be increased according to the priority of data acquisition, thereby adapting to the update rate requirements of different types of sensing data and saving communication bandwidth.
[0062] In some embodiments, after reading the sensing data collected by each sensor in the first sensor group through the first interface in a time-division manner during the first cycle, data processing is performed on the sensing data collected by the position sensor in the first sensor group to obtain the position data of the first finger mechanism during the first cycle.
[0063] In one feasible implementation, the sensing data collected by the position sensors in the first sensor group can be first filtered, such as by removing noise such as electromagnetic interference and mechanical vibration through a low-pass filter or digital filtering algorithm; then the filtered sensing data can be calibrated and standardized, such as by zero-point calibration and temperature compensation; then the calibrated sensing data can be processed by forward or backward kinematics to obtain the position data of the first finger mechanism in the first cycle.
[0064] Similarly, in the second cycle, after reading the sensing data collected by each sensor in the second sensor group through the first interface in a time-division manner, data processing can also be performed on the sensing data collected by the position sensor in the second sensor group to obtain the position data of the first finger mechanism in the second cycle.
[0065] The principle of performing data processing on the sensing data collected by the position sensors in the second sensor group to obtain the position data of the first finger mechanism in the second cycle is the same as the aforementioned principle of performing data processing on the sensing data collected by the position sensors in the first sensor group to obtain the position data of the first finger mechanism in the first cycle.
[0066] The position data of the finger mechanism is obtained by performing data calculations on the position sensor data collected within a cycle, which can improve the real-time performance and accuracy of the finger mechanism's position data.
[0067] In some embodiments, the above Figure 5 The corresponding method also includes the following steps A1-A2:
[0068] A1. During the first cycle, the sensor data collected by each sensor in the third sensor group is read in time-division through the second interface.
[0069] Here, the second interface is the interface for connecting to the sensor within the second finger mechanism. The second finger mechanism is any finger mechanism other than the aforementioned first finger mechanism among the multiple finger mechanisms of a dexterous hand. For example, if the first finger mechanism is the index finger mechanism, then the second finger mechanism can be the thumb finger mechanism, the middle finger finger mechanism, the ring finger finger mechanism, or the little finger finger mechanism.
[0070] The third sensor group includes multiple position sensors and a first tactile sensor within the second finger mechanism. Taking the second finger mechanism as a thumb mechanism as an example, assuming the thumb mechanism includes two position sensors and two tactile sensors, with the two position sensors in the thumb mechanism being position sensor 21 and position sensor 22 respectively, and the two tactile sensors in the thumb mechanism being tactile sensor 21 and tactile sensor 22 respectively, then the third sensor group includes position sensor 21, position sensor 22, and tactile sensor 21; or, the third sensor group includes position sensor 21, position sensor 22, and tactile sensor 22.
[0071] The implementation principle of reading the sensor data collected by each sensor in the third sensor group through the second interface in a time-division manner during the first cycle is the same as the implementation principle of reading the sensor data collected by each sensor in the first sensor group through the first interface in a time-division manner during the first cycle. Please refer to the relevant description of step S301 above, which will not be repeated here.
[0072] A2. Perform data processing on the sensing data collected by the position sensors in the third sensor group to obtain the position data of the second finger mechanism in the first cycle.
[0073] The principle of performing data processing on the sensing data collected by the position sensors in the third sensor group to obtain the position data of the second finger mechanism in the first cycle is the same as the aforementioned principle of performing data processing on the sensing data collected by the position sensors in the first sensor group to obtain the position data of the first finger mechanism in the first cycle.
[0074] In steps A1-A2 above, within the same cycle, sensing data collected by multiple position sensors of other finger mechanisms are also acquired through other interfaces and data processing is performed. This allows the position data of all fingers of the dexterous hand to be obtained within one cycle, which is beneficial for better control of the dexterous hand.
[0075] Understandably, for each finger mechanism of a dexterous hand, the sensing data within the finger mechanism can be obtained through the aforementioned steps S301-S302; and the position data of each finger mechanism within one cycle can be calculated through the aforementioned steps A1-A2 to obtain the position data of each finger mechanism of the dexterous hand within one cycle.
[0076] In some embodiments, after reading the sensing data collected by each sensor in the second sensor group through the first interface in a time-division manner during the second period, the above... Figure 5 The corresponding method also includes the following steps B1-B2:
[0077] B1. Generate a tactile sensing data package based on the sensing data collected by each tactile sensor in multiple finger mechanisms during the first and second cycles.
[0078] The tactile sensing data package contains sensing data collected by various tactile sensors within the finger.
[0079] In one feasible implementation, the tactile sensing data package can be generated through the following steps a1-a2:
[0080] a1. Obtain the first sensor dataset from the sensor data collected by each tactile sensor in the multiple finger mechanisms during the first and second cycles.
[0081] The first sensor dataset includes multiple sensor data belonging to the same finger mechanism, that is, sensor data read through the same interface in multiple cycles. Taking the finger mechanism as an example of the first finger mechanism described in the aforementioned step S301, the first sensor dataset includes sensor data collected by tactile sensor 11, sensor data collected by tactile sensor 12, and sensor data collected by tactile sensor 13.
[0082] a2. Generate a tactile sensing data package based on the interface identifier corresponding to the first sensor dataset, multiple sensing data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensing data in the first sensor dataset.
[0083] The interface identifier is used to uniquely indicate the interface. The interface identifier can be used by the task control system to identify the specific finger mechanism, that is, to determine which finger mechanism the tactile sensing data packet belongs to. The sensor identifier is used to uniquely indicate the sensor. The sensor identifier can be used by the task control system to identify the specific tactile sensor, that is, to determine the tactile state on the specific knuckle.
[0084] Specifically, the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset are added to the same data packet to generate a tactile sensor data packet. That is, the tactile sensor data packet includes the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset.
[0085] For example, if the first sensing dataset includes sensing data collected by tactile sensor 11, tactile sensor 12, and tactile sensor 13, then the interface identifier of the first interface, the sensing data collected by tactile sensor 11, tactile sensor 12, and tactile sensor 13, as well as the sensor identifiers of tactile sensor 11, tactile sensor 12, and tactile sensor 13, can be added to the same data packet to obtain a tactile sensing data packet.
[0086] In steps a1-a2 above, a tactile sensing data package is generated based on the sensing data collected by the tactile sensor of a finger mechanism. This reduces the amount of data in the data package and makes it easier for the task control system to directly perceive the tactile state of each finger mechanism based on each tactile sensing data package.
[0087] In some embodiments, timestamps of multiple sensor data points in the first sensor dataset can be added to the tactile sensor data packet. The timestamps of the multiple sensor data points are used to indicate the acquisition time of the sensor data. Adding the timestamps of the sensor data to the data packet makes it easier for the task control system to identify whether multiple sensor data points in the first sensor dataset belong to tactile sensor data acquired within the same time period (including multiple consecutive cycles), thereby determining whether the sensor data in the tactile sensor data packet is usable.
[0088] In another feasible implementation, the tactile sensing data package can be generated through the following steps b1-b3:
[0089] b1. Obtain the first sensor dataset from the sensor data collected by each tactile sensor in the multiple finger mechanisms during the first and second cycles.
[0090] b2. Generate a verification code based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset.
[0091] In one feasible implementation, a data segment can be formed by combining the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, and generating a cyclic redundancy check code for the data segment to obtain the check code.
[0092] b3. Based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, and the checksum, obtain the tactile sensor data package.
[0093] Specifically, the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, and the checksum can be added to the same data packet to generate a tactile sensor data packet. That is, the tactile sensor data packet includes the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, and the checksum.
[0094] In steps b1-b3 above, a checksum is generated before the data packet is generated, and a tactile sensing data packet is generated based on the checksum and the sensing data so that the task control system can verify the accuracy of the sensing data.
[0095] Understandably, for each finger mechanism of a dexterous hand, the tactile sensor data packet corresponding to the finger mechanism can be generated through the above steps a1-a2 or steps b1-b3, thereby obtaining the tactile sensor data packets corresponding to multiple finger mechanisms of the dexterous hand.
[0096] B2. Send tactile sensor data packets to the task control system.
[0097] Here, the task control system is a system that controls the coordinated movement of each finger mechanism / knuckle structure to perform a task (including but not limited to grasping tasks, assembly tasks, etc.). The task control system can be located inside or outside the dexterous hand.
[0098] In steps B1-B2 above, after obtaining the sensor data collected by the tactile sensors of each finger through multiple cycles, a tactile sensor data packet is generated and sent to the task control system so that the task control system can generate corresponding control tasks or control instructions, thereby realizing the control of the dexterous hand.
[0099] The method of this application has been described above; the apparatus of this application will be described below.
[0100] See Figure 6 , Figure 6 This is a schematic diagram of a data acquisition device for a dexterous hand according to an embodiment of this application. This device is applied to a dexterous hand. For a description of dexterous hands, please refer to... Figure 3 and Figure 4 Corresponding description; such as Figure 6 As shown, the dexterous hand data acquisition device 40 includes:
[0101] The time-division reading module 401 is used to read the sensing data collected by each sensor in the first sensor group through the first interface in the first cycle. The first interface is an interface connected to the sensor in the first finger mechanism. The first finger mechanism is any one of the plurality of finger mechanisms. The first sensor group includes a plurality of position sensors and a first tactile sensor in the first finger mechanism.
[0102] The time-division reading module 402 is further configured to read the sensing data collected by each sensor in the second sensor group through the first interface in a time-division manner during the second cycle, wherein the second cycle is the next cycle after the first cycle, and the second sensor group includes multiple position sensors and a second tactile sensor in the first finger mechanism.
[0103] It should be noted that the aforementioned dexterous hand data acquisition device 40 can execute the dexterous hand data acquisition method provided in the embodiments of this application, and has the corresponding functional modules and beneficial effects of the method. Technical details not described in detail in the embodiments can be found in the dexterous hand data acquisition method provided in the embodiments of this application.
[0104] See Figure 7 , Figure 7 This is a schematic diagram of the structure of a main control chip provided in an embodiment of this application. The main control chip 50 includes a processor 501 and a memory 502. The memory 502 is connected to the processor 501, for example, via a bus.
[0105] Processor 501 is configured to support the main control chip 50 in performing the corresponding functions in the methods described in the above method embodiments. Processor 501 may be a central processing unit (CPU), a network processor (NP), a hardware chip, or any combination thereof. The aforementioned hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The aforementioned PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.
[0106] Memory 502 is used to store program code, etc. Memory 502 may include volatile memory (VM), such as random access memory (RAM); memory 502 may also include non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); memory 502 may also include combinations of the above types of memory.
[0107] The memory 502 is used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the dexterous hand data acquisition method in the embodiments of this application. The processor executes various functional applications and data processing of the dexterous hand data acquisition method by running the non-volatile software programs, instructions, and modules stored in the memory, thereby realizing the functions of the dexterous hand data acquisition method provided in the above method embodiments.
[0108] The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function. The data storage area may store data created based on the use of the dexterous hand data acquisition device. In some embodiments, the memory may include memory remotely located relative to the processor, which can be connected to the dexterous hand data acquisition device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0109] The one or more modules are stored in the memory. When executed by the one or more processors, they perform the dexterous hand data acquisition method in any of the above method embodiments. For example, they perform the method steps described in the above method embodiments to realize the functions of the modules described in the above device embodiments.
[0110] This application also provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the method described in the foregoing embodiments.
[0111] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.
[0112] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.
Claims
1. A method for data acquisition using a dexterous hand, characterized in that, The invention is applied to a dexterous hand, which includes multiple finger mechanisms. Each finger mechanism is equipped with a first tactile sensor, a second tactile sensor, and multiple position sensors. The first tactile sensor, the second tactile sensor, and the multiple position sensors in the same finger mechanism are connected to the same interface. The method includes: During the first cycle, the sensing data collected by each sensor in the first sensor group is read in time-division through the first interface. The first interface is an interface connected to the sensor in the first finger mechanism. The first finger mechanism is any one of the plurality of finger mechanisms. The first sensor group includes a plurality of position sensors and a first tactile sensor in the first finger mechanism. During the second cycle, the sensing data collected by each sensor in the second sensor group is read in a time-division manner through the first interface. The second cycle is the next cycle after the first cycle. The second sensor group includes multiple position sensors and a second tactile sensor in the first finger mechanism.
2. The method according to claim 1, characterized in that, The step of reading the sensing data collected by each sensor in the first sensor group through the first interface in a time-division manner during the first period includes: In response to the target timer interrupt signal, the sensing data collected by each sensor in the first sensor group is read in a time-division manner through the first interface, wherein the target timer interrupt signal is the timer interrupt signal with the highest priority.
3. The method according to claim 1, characterized in that, After reading the sensor data collected by each sensor in the first sensor group through the first interface in a time-division manner during the first period, the method further includes: Data processing is performed on the sensing data collected by the position sensors in the first sensor group to obtain the position data of the first finger mechanism within the first cycle.
4. The method according to claim 1, characterized in that, After reading the sensor data collected by each sensor in the second sensor group through the first interface in a time-division manner during the second period, the method further includes: A tactile sensing data package is generated based on the sensing data collected by each tactile sensor in the plurality of finger mechanisms during the first cycle and the second cycle. The tactile sensing data packet is sent to the task control system.
5. The method according to claim 4, characterized in that, The step of generating a tactile sensing data packet based on the sensing data collected by each tactile sensor in the plurality of finger mechanisms during the first cycle and the second cycle includes: In the first cycle and the second cycle, a first sensor dataset is obtained from the sensor data collected by each tactile sensor in the plurality of finger mechanisms. The first sensor dataset includes multiple sensor data belonging to the same finger mechanism. The tactile sensing data package is generated based on the interface identifier corresponding to the first sensor dataset, multiple sensing data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensing data in the first sensor dataset.
6. The method according to claim 5, characterized in that, Before generating the tactile sensing data packet based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, the method further includes: A verification code is generated based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset. The step of generating the tactile sensing data package based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, and the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset includes: The tactile sensing data package is generated based on the interface identifier corresponding to the first sensor dataset, multiple sensor data in the first sensor dataset, the sensor identifiers corresponding to each of the multiple sensor data in the first sensor dataset, and the verification code.
7. The method according to claim 1, characterized in that, The method further includes: During the first cycle, the sensing data collected by each sensor in the third sensor group is read in time-division through the second interface. The second interface is an interface connected to the sensor in the second finger mechanism. The second finger mechanism is any finger mechanism other than the first finger mechanism among the plurality of finger mechanisms. The third sensor group includes a plurality of position sensors and a first tactile sensor in the second finger mechanism. Data processing is performed on the sensing data collected by the position sensors in the third sensor group to obtain the position data of the second finger mechanism within the first cycle.
8. A dexterous hand data acquisition device, characterized in that, The invention is applied to a dexterous hand, which includes multiple finger mechanisms. Each finger mechanism is equipped with a first tactile sensor, a second tactile sensor, and multiple position sensors. The first tactile sensor, the second tactile sensor, and the multiple position sensors in the same finger mechanism are connected to the same interface. The device includes: The time-division reading module is used to read the sensor data collected by each sensor in the first sensor group in a time-division manner through the first interface during the first cycle. The first interface is an interface connected to the sensor in the first finger mechanism. The first finger mechanism is any one of the plurality of finger mechanisms. The first sensor group includes a plurality of position sensors and a first tactile sensor in the first finger mechanism. The time-division reading module is also used to read the sensing data collected by each sensor in the second sensor group through the first interface in a time-division manner during the second cycle, wherein the second cycle is the next cycle after the first cycle, and the second sensor group includes multiple position sensors and a second tactile sensor in the first finger mechanism.
9. A main control chip, characterized in that, The system includes a memory and a processor, the memory being connected to the processor, the processor being configured to execute one or more computer programs stored in the memory, and the processor, when executing the one or more computer programs, causing the main control chip to perform the method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1-7.