CN112976025B - Object soft and hard recognition and self-adaptive grabbing method based on humanoid manipulator device - Google Patents
Object soft and hard recognition and self-adaptive grabbing method based on humanoid manipulator device Download PDFInfo
- Publication number
- CN112976025B CN112976025B CN202110199135.5A CN202110199135A CN112976025B CN 112976025 B CN112976025 B CN 112976025B CN 202110199135 A CN202110199135 A CN 202110199135A CN 112976025 B CN112976025 B CN 112976025B
- Authority
- CN
- China
- Prior art keywords
- hardness
- finger
- forefinger
- grasped object
- electric push
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J15/00—Gripping heads and other end effectors
- B25J15/0009—Gripping heads and other end effectors comprising multi-articulated fingers, e.g. resembling a human hand
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
Abstract
The invention discloses an object hardness identification and self-adaptive grabbing method based on a humanoid manipulator device. The method for realizing the soft and hard identification and the self-adaptive grabbing of the object comprises the steps of setting the output force of a miniature electric push rod; judging the hardness of the grasped object; the device comprises a bracket, a humanoid manipulator, a sensing module, a control system and a driving module; the human-simulated manipulator is fixed on the support, the sensing module is arranged on fingers of the human-simulated manipulator, the driving module is connected to the back face of a palm of the human-simulated manipulator, the input end of the control system is electrically connected with the sensing module, and the output end of the control system is electrically connected with the driving module. The humanoid manipulator device can identify the hardness and softness of an object and can adaptively and stably grab the object with different hardness and softness attributes, is constructed by 3D printing and is provided with various sensing and sensing modules, and is expected to become a portable artificial limb or be applied to the field of service robots.
Description
Technical Field
The invention relates to the technical field of humanoid robots, in particular to a method for recognizing hardness and self-adaptive grabbing of an object based on a humanoid manipulator device, which can actively sense the hardness and softness of the object and realize self-adaptive grabbing of the object on the premise of not damaging the object.
Background
With the rapid development of the robot technology, the use of a mechanical gripper to grab and control an object is a basic task in life, production and scientific research, and the hardness and softness as one of the important attributes of the object can influence the grabbing control of the manipulator to the object to a great extent. At present, most of bionic mechanical arm researches are on the structure and the driving mode of the mechanical arm, and the mechanical arm can complete simple stretching, grabbing and other actions like a human hand, but the most defect of the mechanical arm researches is that multi-mode tactile information related to touch, force sense, pressure sense and the like cannot be acquired like the human hand. In the contact process of the manipulator and the object, the touch feedback information is a dynamic variable quantity, and the soft and hard attributes of the object cannot be well identified by only depending on the force feedback signal.
From the requirement of realizing the self functions of the bionic manipulator, most objects in real life are rigid or deformable, and the acquisition of the soft and hard attributes of the objects by sensing in advance is a precondition and an important guarantee for realizing the stable grabbing of the objects by the manipulator. When the manipulator grabs a hard object, the manipulator can control the magnitude of the grabbing force, so that the object can be stably grabbed, and the damage to the structure of the manipulator caused by excessive force in the grabbing process can be prevented; when the soft object is grabbed, the manipulator can be ensured to grab the object with proper grabbing force, and the grabbed object is prevented from being deformed or damaged due to overlarge grabbing force. The manipulator applies proper grabbing force according to the soft and hard characteristics of the object, so that the grabbed object is not damaged by extrusion, and the manipulator does not bear excessive reaction force, thereby realizing stable and flexible grabbing action.
Disclosure of Invention
The invention provides an object hardness identification and self-adaptive grabbing method based on a humanoid manipulator device, aiming at the defects that the bionic manipulator can not accurately reflect the hardness and softness of an object only by relying on a force feedback signal.
The technical scheme of the invention is as follows:
humanoid manipulator device
The device comprises a bracket, a humanoid manipulator, a sensing module, a control system and a driving module; the human-simulated manipulator is fixed on the support, the sensing module is arranged on fingers of the human-simulated manipulator, the driving module is connected to the back face of a palm of the human-simulated manipulator, the input end of the control system is electrically connected with the sensing module, and the output end of the control system is electrically connected with the driving module.
The human-simulated manipulator comprises a palm and five fingers connected with the palm, wherein the five fingers comprise an index finger, a middle finger, a ring finger, a thumb and a little finger; the sensing module comprises a liquid metal flexible strain sensor and a PVDF piezoelectric sensor; the control system comprises an STM32 control panel and an upper computer; the driving module comprises three miniature electric push rods of a first miniature electric push rod, a second miniature electric push rod and a third miniature electric push rod, a torsion spring and a driving flexible rope;
PVDF piezoelectric sensors wrapped by Polydimethylsiloxane (PDMS) are pasted on fingertips of an index finger, a ring finger and a thumb of the humanoid manipulator, liquid metal flexible strain sensors are pasted on a distal knuckle rotating joint and a middle knuckle rotating joint of the index finger of the humanoid manipulator, torsion springs are uniformly distributed at the distal knuckle rotating joints of the index finger, the middle finger, the ring finger and the little finger respectively, one torsion spring is uniformly distributed at the middle knuckle joints of the index finger, the middle finger, the ring finger and the little finger respectively, one torsion spring is distributed at the distal knuckle rotating joint of the thumb, no proximal knuckle rotating joint exists in five fingers of the humanoid manipulator, and proximal knuckles of the five fingers of the humanoid manipulator are vertically fixed on a palm and do not bend; the three miniature electric push rods are vertically fixed on the back of the palm, in the specific implementation, the bottoms of the three miniature electric push rods are vertically fixed on the bottom of the back of the palm, the tops of the three miniature electric push rods are fixed in the middle of the back of the palm, the three miniature electric push rods realize the bending of fingers by stretching the driving flexible ropes, one end of each driving flexible rope is fixed on the far knuckle of each finger of an index finger, a ring finger, a middle finger, a big finger and a small finger, and the other end of each driving flexible rope sequentially penetrates through the far knuckle, the far knuckle rotating joint, the middle knuckle rotating joint and the near knuckle of each finger and then is tied to the miniature electric push rods on the back of the palm;
PVDF piezoelectric sensor is through AD620 filtering collection module enlargies and connect the analog input end of STM32 control panel behind the conversion filtering, the flexible strain transducer of liquid metal is through dividing another input of electric road roller electricity connection STM32 control panel, the signal unidirectional transfer of PVDF piezoelectric sensor and the flexible strain transducer of liquid metal is for the STM32 control panel, the host computer is connected to an output electric connection of STM32 control panel, STM32 control panel and the two-way transmission signal of host computer, three miniature electric putter is connected to another output electricity of STM32 control panel, miniature electric putter realizes crooked and extension through the flexible rope of tensile drive and the imitative people manipulator of torsional spring control.
The upper computer and the STM32 control board control the three miniature electric push rods to contract, so that the driving flexible ropes of the fingers are tensioned, the torsion springs at the knuckle rotating joints of the fingers are bent, and the fingers are further driven to bend; the upper computer and the STM32 control board are stretched through controlling three miniature electric push rods, and then the drive flexible rope of each finger is loosened to make the drive flexible rope loosened, and the extension of each finger is realized under the effect of the torsional spring of each knuckle rotation joint department to each finger.
The drive flexible rope on the forefinger is tied on a first miniature electric push rod close to the back of the palm of the thumb, the drive flexible ropes on the middle finger, the ring finger and the little finger are all tied on a second miniature electric push rod in the middle of the back of the palm, and the drive flexible rope on the little finger is tied on a third miniature electric push rod far away from the back of the palm of the thumb.
The PVDF piezoelectric sensor monitors dynamic force, in the invention, when the pressure of a grasped object received by a finger changes, the PVDF piezoelectric sensor generates a voltage signal, and when the pressure of the grasped object received by the finger does not change, the PVDF piezoelectric sensor does not generate the voltage signal.
The PVDF piezoelectric sensor is prepared by a salivation method, the PVDF piezoelectric sensor is wrapped by Polydimethylsiloxane (PDMS) through a layered mode inversion method and is solidified in a 3D printed template, and the surface of a composite structure obtained after solidification has a fingerprint microstructure similar to human skin, so that the detection sensitivity is improved, the friction of a contact surface is increased, and the sliding during grabbing is avoided.
The liquid metal flexible strain sensor is prepared by adopting a mask method, a snake-shaped circuit structure is arranged in the liquid metal flexible strain sensor to increase the sensitivity of the liquid metal flexible strain sensor, and two ends of the liquid metal flexible strain sensor are respectively stuck and fixed at two rotating joints of an index finger in a sticking mode through a resin type silicon rubber special strong adhesive 988.
Second, object hardness identification and self-adaptive grabbing method based on humanoid manipulator device
The method comprises the following specific steps:
step S1: setting the output force of the miniature electric push rod:
step S11: firstly, the upper computer sends a program instruction with a PWM duty ratio signal of 80 to an STM32 control board, the STM32 control board controls a first micro electric push rod to retract, further pulls a driving flexible rope on a forefinger to enable the driving flexible rope to be tensioned, a torsional spring on the forefinger bends to enable the forefinger to bend, the forefinger bends to contact with a grasped object and completes knocking, when the forefinger does not contact with the grasped object, no voltage signal is generated by a PVDF piezoelectric sensor on the forefinger, when the forefinger contacts with the grasped object, a voltage signal is generated by the PVDF piezoelectric sensor on the forefinger, when the forefinger completes knocking, the grasped object receives a feedback force given by a humanoid manipulator, the PVDF piezoelectric sensor on the forefinger senses the grasped object and generates a voltage signal, after the forefinger completes knocking, the upper computer sends the program instruction with the PWM duty ratio signal of-80 to the STM32 control board, the STM32 control board controls the first micro electric push rod to extend, thereby driving the flexible driving rope on the forefinger to be loosened, and the torsion spring on the forefinger deforms and restores and drives the forefinger to restore to a straight state;
step S12: the voltage signal that PVDF piezoelectric sensor on the forefinger produced amplifies and converts the post-filtering to voltage signal output and sends to STM32 control panel through AD620 filtering collection module, judges the voltage interval that voltage signal belongs to through STM32 control panel operation processing, then STM32 control panel obtains the hardness grade that the voltage interval corresponds according to the following corresponding relation according to the voltage interval that voltage signal belongs to and obtains the hardness grade of being grabbed the object to the hardness grade that will be grabbed the object sends for the host computer:
if the voltage interval in which the voltage signal is located is [ 5-10) mv, the hardness grade of the grasped object is A, if the voltage interval in which the voltage signal is located is [ 10-25) mv, the hardness grade of the grasped object is B, if the voltage interval in which the voltage signal is located is [ 25-60) mv, the hardness grade of the grasped object is C, and if the voltage interval in which the voltage signal is located is [ 60-150 ] mv, the hardness grade of the grasped object is D;
the hardness scale a indicates softness, the hardness scale B indicates relatively softness, the hardness scale C indicates somewhat softness, and the hardness scale D indicates relatively hardness.
Step S13: the upper computer judges a PWM duty ratio signal interval corresponding to the hardness grade of the grasped object and an output force interval corresponding to the PWM duty ratio signal interval according to the obtained hardness grade of the grasped object and the following corresponding relation:
if the hardness grade of the grasped object is A, a PWM duty cycle signal interval corresponding to the hardness grade A is [ 80-100 ], and an output force interval corresponding to the PWM duty cycle signal interval [ 80-100) is [ 2-2.5) N; if the hardness grade of the grasped object is B, the PWM duty cycle signal interval corresponding to the hardness grade B is [ 100-140 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 100-140 ] is [ 2.5-5) N; if the hardness grade of the grasped object is C, the PWM duty cycle signal interval corresponding to the hardness grade C is [ 140-190 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 140-190) is [ 5-8.5) N; if the hardness grade of the grasped object is D, the PWM duty cycle signal interval corresponding to the hardness grade D is [ 190-255 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 190-255 ] is [ 8.5-12.5 ] N;
the upper computer sends a program instruction of a minimum signal value corresponding to a PWM duty cycle signal interval corresponding to the hardness grade of the grasped object to an STM32 control board according to the obtained hardness grade of the grasped object, and the STM32 control board controls the micro electric push rod to output an output force corresponding to the minimum signal value corresponding to the PWM duty cycle signal interval according to the obtained program instruction;
step S2: judging the hardness of the grasped object:
step S21: the STM32 control panel is with the program command control first miniature electric putter of the minimum signal value that the hardness grade that obtains corresponds that PWM duty cycle signal interval corresponds that step S13 is contracted, and then the drive flexible rope on the taut forefinger makes the drive flexible rope by the tensioning, and the torsional spring on the forefinger is crooked and then makes the forefinger crooked until the forefinger is no longer crooked, forefinger and the object contact of being grabbed and accomplish the press of forefinger to the object of being grabbed, when the forefinger is accomplished to press the object of being grabbed, what the forefinger of imitative people 'S manipulator received is grabbed the object' S that the forefinger receivedPressure F2The output force is equal to the output force corresponding to the minimum signal value corresponding to the PWM duty ratio signal interval output by the first miniature electric push rod;
step S22: in the process of bending the forefinger, the two liquid metal sensors on the forefinger are stretched, the two liquid metal sensors output respective resistance signals, the forefinger stops bending until the resistance values of the resistance signals output by the two liquid metal sensors are not changed any more, the respective resistance signals output by the two liquid metal strain sensors are sent to an STM32 control panel through respective voltage division circuits, the STM32 control panel obtains the two resistance values and sends the two resistance values to an upper computer, the upper computer converts the two resistance values into two corners through analysis and calculation, the lengths of a far knuckle and a middle knuckle of the forefinger are known, and then the upper computer further obtains the fingertip displacement of the forefinger through calculation according to the two corners;
step S23: the hardness sensitivity coefficient S is used as the measurement of hardness of the grasped object, and the displacement of the fingertip of the index finger is equal to the deformation x of the grasped object under pressure in the process that the index finger presses the grasped object2(ii) a The soft and hard sensitivity coefficient S is calculated using the following formula:
in the formula, K1Denotes the elastic coefficient, x, of the index finger tip2Representing the amount of compressive deformation of the gripped object, F2Indicating the pressure applied by the index finger of the manipulator when the pressing is finished;
obtaining the softness and hardness of the grasped object after obtaining the softness and hardness sensitivity coefficient S; the closer the value of the soft and hard sensitivity coefficient S is to 1, the softer the object is; the closer the value of the soft and hard sensitivity coefficient S is to 0, the harder the object is;
step S3: stably grabbing a grabbed object:
step S31: when the grabbed object is grabbed, the mechanical arm is fixed on the humanoid manipulator to realize that the grabbed object is grabbed after five fingers are contacted with the grabbed object; the STM32 control board controls the three micro electric push rods to contract according to the program instruction of the minimum signal value corresponding to the PWM duty ratio signal interval corresponding to the hardness grade obtained in the step S13, so that five fingers including the index finger, the ring finger, the middle finger, the thumb and the small finger of the manipulator are bent;
step S32: when the five fingers bend to be in contact with a grasped object, the PVDF piezoelectric sensors on the index finger, the ring finger and the thumb all generate voltage signals, and when the pressure of the grasped object on the five fingers is equal to the output force corresponding to the minimum signal value corresponding to the PWM duty ratio signal interval output by the three miniature electric push rods, the grasping of the grasped object by the five fingers is completed;
step S33: after the five fingers complete the grabbing of the grabbed object, the PVDF piezoelectric sensors on the index finger, the ring finger and the thumb do not generate voltage signals, and in the grabbing process of the grabbed object under the action of the mechanical arm: friction force is generated between a grasped object and five fingers, if the value of the maximum friction force is larger than or equal to the gravity of the grasped object, relative sliding does not occur between the five fingers and the grasped object, no voltage signal is generated by the PVDF piezoelectric sensors on the index finger, the ring finger and the thumb, the five fingers can stably grasp the grasped object, if the value of the maximum friction force is smaller than the gravity of the grasped object, relative sliding occurs between the five fingers and the grasped object, voltage signals are generated by the PVDF piezoelectric sensors on the index finger, the ring finger and the thumb, and the five fingers do not stably grasp the grasped object;
if no voltage signal is generated by the PVDF piezoelectric sensors on the index finger, the ring finger and the thumb, which indicates that no sliding occurs between the five fingers and the grasped object, the stable grasping of the grasped object by the five fingers is completed;
if at least one of the PVDF piezoelectric sensors on the index finger, the ring finger and the thumb is detected to generate a voltage signal, the sliding between the five fingers and the grasped object is indicated, and then the step S34 is carried out;
step S34: voltage signals generated by the PVDF piezoelectric sensor are amplified, converted and filtered by an AD620 filtering and collecting module, then the voltage signals are output and sent to an STM32 control panel, then the STM32 control panel sends the obtained voltage signals to an upper computer, the upper computer gradually updates and enlarges signal values corresponding to PWM duty cycle signal intervals corresponding to the hardness grade of a grasped object in an instruction program and sends the signal values to an STM32 control panel, the STM32 control panel controls three miniature electric push rods to contract according to the instruction program which is gradually updated and enlarged each time, and then five fingers of a forefinger, a ring finger, a middle finger, a thumb and a little finger of the manipulator are bent;
step S35: repeating the steps S32-S34, and in the step S34, if the adjusted PWM duty ratio signal value reaches the maximum signal value corresponding to the PWM duty ratio signal interval corresponding to the hardness grade of the grasped object, detecting that a voltage signal is generated by the PVDF piezoelectric sensor in the process of grasping the grasped object under the action of the mechanical arm in the step S33, preventing the grasped object from being damaged, directly abandoning the grasping, and judging that the grasping fails, which indicates that five fingers do not finish stably grasping the grasped object.
The invention can be used for judging the hardness of the grasped object, and can also be used for judging the hardness of the grasped object and grasping the grasped object; only the hardness of the grasped object is judged by performing the steps S1 to S2.
The invention has the beneficial effects that:
(1) the human-simulated manipulator system is based on a rigid manipulator framework, but is not limited to a rigid manipulator, and the manipulator sensor is flexible, so that the system for realizing object hardness perception can be applied to a flexible manipulator;
(2) through the PVDF piezoelectric sensor and the liquid metal flexible sensor, the soft and hard of the object are represented and identified through the voltage and strain dual signals, the soft and hard attributes of the object can be accurately identified, and the self-adaptive stable grabbing of the humanoid manipulator for the objects with different soft and hard attributes is realized;
(3) in the grabbing process, whether an object slides can be detected in real time through whether the PVDF piezoelectric sensor outputs a piezoelectric signal or not, and convenience and accuracy are achieved;
(4) the artificial limb is expected to become a portable artificial limb or be applied to the field of service robots by adopting the structural design of a human palm imitation and being provided with various sensing modules;
(5) the manipulator is constructed through 3D printing, so that the manufacturing cost is low and the period is short.
Summarizing, the humanoid manipulator device can identify the hardness and softness of objects and can adaptively and stably grab the objects with different hardness and softness attributes, and the humanoid manipulator device is constructed by 3D printing and is provided with various perception sensing modules, so that the humanoid manipulator device is expected to become a portable artificial limb or be applied to the field of service robots.
Drawings
FIG. 1 is a schematic structural view of a humanoid manipulator of the present invention;
FIG. 2 is a schematic diagram of the index finger of the humanoid manipulator of the present invention;
FIG. 3 is a schematic view of a driving structure of the humanoid manipulator of the present invention;
FIG. 4 is a schematic view of the humanoid robot system of the present invention;
FIG. 5 is a flow chart of the soft and hard identification of the humanoid manipulator system of the invention;
FIG. 6 is a flow chart of the stabilized grasping of the humanoid manipulator system of the present invention;
FIG. 7 is a graph of voltage signals from PVDF voltage sensors as a robot recognizes objects of different stiffness in accordance with the present invention;
FIG. 8 is a signal diagram of a liquid metal strain sensor for identifying an object by a manipulator according to the present invention; in FIG. 8, R1 represents the measured signal of the flexible liquid metal strain sensor at the knuckle of the index finger; r2 represents the measured signal of the flexible liquid metal strain sensor at the distal knuckle revolute joint of the index finger.
In the figure, 1 forefinger, 2 middle finger, 3 ring finger, 4 little finger, 6 thumb, 7 palm, 8 supports, 9PVDF piezoelectric sensors, 10 torsion springs, 11 distal knuckle rotating joints, 12 driving flexible ropes, 13 liquid metal flexible strain sensors, 15 middle knuckle rotating joints, 16 first miniature electric push rods, 17 second miniature electric push rods and 18 third miniature electric push rods.
Detailed Description
In order to more clearly explain the technical solution of the present invention, the following detailed description will be made with reference to the accompanying drawings and examples.
As shown in fig. 1, the device of the invention comprises a bracket 8, a humanoid manipulator, a sensing module, a control system and a driving module; the humanoid manipulator is fixed on the support 8, the sensing module is arranged on fingers of the humanoid manipulator, the driving module is arranged and connected to the back face of a palm of the humanoid manipulator, the input end of the control system is electrically connected with the sensing module, and the output end of the control system is electrically connected with the driving module.
As shown in fig. 2 and 3, the humanoid manipulator comprises a palm 7 and five fingers connected with the palm 7, wherein the five fingers comprise an index finger 1, a middle finger 2, a ring finger 3, a big finger 6 and a small finger 4; the sensing module comprises a liquid metal flexible strain sensor 13 and a PVDF piezoelectric sensor 9; the control system comprises an STM32 control panel and an upper computer; the driving module comprises three miniature electric push rods of a first miniature electric push rod 16, a second miniature electric push rod 17 and a third miniature electric push rod 18, a torsion spring 10 and a driving flexible rope 12;
PVDF piezoelectric sensors 9 wrapped by polydimethylsiloxane PDMS are pasted on fingertips of an index finger 1, a ring finger 3 and a thumb 6 of the human-simulated manipulator, liquid metal flexible strain sensors 13 are pasted on a distal knuckle rotating joint 11 and a middle knuckle rotating joint 15 of the index finger 1 of the human-simulated manipulator, torsion springs 10 are respectively arranged at distal knuckle rotating joints of the index finger 1, a middle finger 2, the ring finger 3 and the small finger 4, torsion springs 10 are respectively arranged at middle knuckle joints of the index finger 1, the middle finger 2, the ring finger 3 and the small finger 4, torsion springs 10 are respectively arranged at distal knuckle rotating joints of the thumb 6, no proximal knuckle rotating joint exists in five fingers of the human-simulated manipulator, and proximal knuckles of the five fingers of the human-simulated manipulator are vertically fixed on a palm 7 and do not bend; the three miniature electric push rods are vertically fixed on the back of the palm 7, in the specific implementation, the bottoms of the three miniature electric push rods are vertically fixed on the bottom of the back of the palm, the tops of the three miniature electric push rods are fixed in the middle of the back of the palm 7, the three miniature electric push rods realize the bending of fingers by stretching the driving flexible ropes 12, one end of each driving flexible rope 12 is fixed on the far knuckle of each finger of the forefinger 1, the ring finger 3, the middle finger 2, the big finger 6 and the small finger 4, and the other end of each driving flexible rope 12 sequentially penetrates through the far knuckle, the far knuckle rotating joint, the middle knuckle rotating joint and the near knuckle of each finger and then is tied to the miniature electric push rods on the back of the palm 7;
PVDF piezoelectric sensor 9 is through AD620 filtering collection module enlargies and connect the analog input end of STM32 control panel behind the conversion filtering, flexible strain transducer 13 of liquid metal is through dividing another input of electric road roller electricity connection STM32 control panel, the signal unidirectional transfer of PVDF piezoelectric sensor 9 and flexible strain transducer 13 of liquid metal is for the STM32 control panel, the host computer is connected to an output electricity of STM32 control panel, STM32 control panel and the two-way transmission signal of host computer, three miniature electric putter is connected to another output electricity of STM32 control panel, miniature electric putter realizes crooked and extension through tensile drive flexible rope 12 and torsional spring 10 control humanoid manipulator.
As shown in fig. 4, the upper computer and the STM32 control board control the three micro electric push rods to contract, and further tension the driving flexible ropes 12 of each finger, so that the driving flexible ropes 12 are tensioned, and the torsion springs 10 at the knuckle rotating joints of each finger are bent, and further the fingers are driven to bend; the upper computer and the STM32 control board control the three micro electric push rods to stretch, and further loosen the driving flexible ropes 12 of each finger to enable the driving flexible ropes 12 to be loosened, and the extension of each finger is realized under the action of the torsion springs 10 at the rotating joints of each knuckle.
The driving flexible rope 12 on the forefinger 1 is tied on a first miniature electric push rod 16 close to the back of the palm 7 of the thumb, the driving flexible ropes 12 on the middle finger 2, the ring finger 3 and the little finger 4 are all tied on a second miniature electric push rod 17 in the middle of the back of the palm 7, and the driving flexible rope 12 on the little finger 6 is tied on a third miniature electric push rod 18 far away from the back of the palm 7 of the thumb.
The PVDF piezoelectric transducer 9 of the invention monitors dynamic forces, when the force changes, the PVDF piezoelectric transducer 9 generates a voltage signal, when the force does not change, the PVDF piezoelectric transducer 9 does not generate a voltage signal.
The PVDF piezoelectric sensor 9 is prepared by a salivation method, the PVDF piezoelectric sensor 9 is wrapped by polydimethylsiloxane PDMS through a layered mode inversion method and is solidified in a 3D printed template, and the surface of a composite structure obtained after solidification has a fingerprint microstructure similar to human skin, so that the detection sensitivity is improved, the friction of a contact surface is increased, and the sliding during grabbing is avoided.
The liquid metal flexible strain sensor 13 is prepared by adopting a mask method, a snake-shaped circuit structure is arranged in the liquid metal flexible strain sensor 13 to increase the sensitivity of the liquid metal flexible strain sensor, and two ends of the liquid metal flexible strain sensor 13 are respectively stuck and fixed at two rotating joints of the forefinger 1 in a sticking way through a resin type silicon rubber special strong adhesive 988.
As shown in fig. 5 and 6, the method for recognizing hardness of an object and performing adaptive grasping by using a humanoid manipulator device for recognizing hardness of the object and performing adaptive grasping includes the following steps:
step S1: setting and acquiring the output force of the miniature electric push rod:
step S11: firstly, the upper computer sends a program instruction with a PWM duty ratio signal of 80 to an STM32 control board, the STM32 control board controls a first micro electric push rod 16 to retract, and further pulls a driving flexible rope 12 on a forefinger 1 to tension the driving flexible rope 12, a torsion spring 10 on the forefinger 1 bends to bend the forefinger 1, the forefinger 1 bends to contact with a grasped object and complete knocking, when the forefinger 1 does not contact with the grasped object, no voltage signal is generated by a PVDF piezoelectric sensor 9 on the forefinger 1, when the forefinger 1 contacts with the grasped object, a voltage signal is generated by the PVDF piezoelectric sensor 9 on the forefinger 1, when the forefinger 1 completes knocking, the grasped object receives a feedback force given by a humanoid manipulator, the PVDF piezoelectric sensor 9 on the forefinger 1 senses the grasped object and generates the voltage signal, after the forefinger 1 completes knocking, the upper computer sends the program instruction with the PWM duty ratio signal of-80 to the STM32 control board, the STM32 control panel controls the first miniature electric push rod 16 to extend out, and then drives the driving flexible rope 12 on the forefinger 1 to be loosened, the torsion spring 10 on the forefinger 1 deforms and restores and drives the forefinger 1 to restore to the straight state;
step S12: the voltage signal that PVDF piezoelectric sensor 9 on forefinger 1 produced amplifies and converts the post-filtration through AD620 filtering collection module and sends voltage signal output to STM32 control panel, judges the voltage interval that voltage signal belongs to through STM32 control panel operation processing, then the hardness grade that obtains the hardness grade that the voltage interval corresponds according to the voltage interval that voltage signal belongs to the STM32 control panel according to the following corresponding relation obtains the hardness grade of being grabbed the object, and will be grabbed the hardness grade of object and send for the host computer:
if the voltage interval in which the voltage signal is located is [ 5-10) mv, the hardness grade of the grasped object is A, if the voltage interval in which the voltage signal is located is [ 10-25) mv, the hardness grade of the grasped object is B, if the voltage interval in which the voltage signal is located is [ 25-60) mv, the hardness grade of the grasped object is C, and if the voltage interval in which the voltage signal is located is [ 60-150 ] mv, the hardness grade of the grasped object is D;
the hardness scale a indicates softness, the hardness scale B indicates relatively softness, the hardness scale C indicates somewhat softness, and the hardness scale D indicates relatively hardness.
Step S13: the upper computer judges a PWM duty ratio signal interval corresponding to the hardness grade of the grasped object and an output force interval corresponding to the PWM duty ratio signal interval according to the obtained hardness grade of the grasped object and the following corresponding relation:
if the hardness grade of the grasped object is A, the PWM duty cycle signal interval corresponding to the hardness grade A is [ 80-100 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 80-100) is [ 2-2.5N; if the hardness grade of the grasped object is B, the PWM duty cycle signal interval corresponding to the hardness grade B is [ 100-140 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 100-140 ] is [ 2.5-5N; if the hardness grade of the grasped object is C, the PWM duty cycle signal interval corresponding to the hardness grade C is [ 140-190 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 140-190) is [ 5-8.5N; if the hardness grade of the grasped object is D, the PWM duty cycle signal interval corresponding to the hardness grade D is [ 190-255 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 190-255 ] is [ 8.5-12.5 ] N;
the upper computer sends a program instruction of a minimum signal value corresponding to a PWM duty cycle signal interval corresponding to the hardness grade of the grasped object to an STM32 control board according to the obtained hardness grade of the grasped object, and the STM32 control board controls the micro electric push rod to output an output force corresponding to the minimum signal value corresponding to the PWM duty cycle signal interval according to the obtained program instruction;
step S2: judging the hardness of the grasped object:
step S21: the STM32 control board controls the first micro electric push rod 16 to contract according to the program instruction of the minimum signal value corresponding to the PWM duty cycle signal interval corresponding to the hardness grade obtained in the step S13, further, the driving flexible rope 12 on the index finger 1 is tensioned, the torsion spring 10 on the index finger 1 bends, further, the index finger 1 bends until the index finger 1 does not bend, the index finger 1 contacts with the grasped object and completes the pressing of the index finger 1 on the grasped object, and when the index finger 1 finishes the pressing of the grasped object, the pressure F of the grasped object received by the index finger 1 of the humanoid manipulator2Is equal to the magnitude of the output force corresponding to the minimum signal value corresponding to the PWM duty cycle signal interval output by the first miniature electric push rod 16;
step S22: in the bending process of the index finger 1, the two liquid metal sensors 13 on the index finger 1 are stretched, the two liquid metal sensors 13 output respective resistance signals, the index finger 1 stops bending when the resistance values of the resistance signals output by the two liquid metal sensors 13 are not changed, the respective resistance signals output by the two liquid metal strain sensors 13 are sent to an STM32 control board through respective voltage division circuits, the STM32 control board obtains two resistance values and sends the two resistance values to an upper computer, the upper computer converts the two resistance values into two rotation angles through analysis and calculation, and as shown in FIG. 8, R1 represents the measured signals of the flexible liquid metal strain sensors at the knuckle rotating joint of the index finger; r2 represents the measured signal of the flexible liquid metal strain sensor at the rotating joint of the far knuckle of the index finger, the length of the far knuckle and the middle knuckle of the index finger is known, and then the upper computer further calculates to obtain the fingertip displacement of the index finger 1 according to the two rotating angles;
step S23: to be softThe hard sensitivity coefficient S is used as the measurement of the hardness of the grasped object, and the displacement of the fingertip of the index finger 1 is equal to the deformation x of the grasped object under pressure in the process that the index finger 1 presses the grasped object2(ii) a The soft and hard sensitivity coefficient S is calculated using the following formula:
in the formula, K1Denotes the elastic coefficient, x, of the index finger 1 tip2Representing the amount of compressive deformation of the gripped object, F2Indicating the pressure applied by the index finger 1 of the manipulator when the pressing is finished;
obtaining the softness and hardness of the grasped object after obtaining the softness and hardness sensitivity coefficient S; the closer the value of the soft and hard sensitivity coefficient S is to 1, the softer the object is; the closer the value of the soft and hard sensitivity coefficient S is to 0, the harder the object is;
in specific implementation, the mechanical arm is regarded as elastic deformation, and under a certain mechanical pressure F, the overall deformation of the system can be expressed as the deformation x of the fingertip sensor1And deformation x of the object to be grasped2Then the equivalent elastic coefficient of the system can be expressed as:
assuming that the system elastic coefficient varies with contact, S ═ dK/dK is defined2Is the soft and hard sensitivity coefficient. Wherein, dK2Representing the elastic coefficient of the gripped object.
If the contact force is low, the mechanical behavior of the grasped object and the fingertip sensor is considered to be in accordance with linear elastic deformation, and K is used at the moment2As variables and derived from them, we can:
the elastic coefficient of the grabbed object is obtained through the compressive deformation quantity of the object, and then the soft and hard sensitivity coefficient S capable of representing the soft and hard attributes of the object is calculated, so that the humanoid manipulator system can sense the soft and hard attributes of the object.
After the soft and hard attributes of the object are obtained, the humanoid manipulator system can apply different grabbing forces to the object with different hardness, and then the manipulator can grab the object in a self-adaptive mode.
The humanoid manipulator can also judge whether preslip and slide occur in the grabbing process through the signal output of the piezoelectric sensor, and further guarantee that the manipulator stably grabs the object.
Step S3: stably grabbing a grabbed object:
step S31: when the grabbed object is grabbed, the mechanical arm is fixed on the humanoid manipulator to realize that the grabbed object is grabbed after five fingers are contacted with the grabbed object; the STM32 control board controls the three micro electric push rods to contract according to the program instruction of the minimum signal value corresponding to the PWM duty ratio signal interval corresponding to the hardness grade obtained in the step S13, so that five fingers of the index finger 1, the ring finger 3, the middle finger 2, the thumb 6 and the little finger 4 of the manipulator are bent;
step S32: when the five fingers bend to be in contact with a grasped object, the PVDF piezoelectric sensors 9 on the index finger 1, the ring finger 3 and the thumb 6 all generate voltage signals, and when the pressure of the grasped object on the five fingers is equal to the output force corresponding to the minimum signal value corresponding to the PWM duty cycle signal interval output by the three miniature electric push rods, the grasping of the grasped object by the five fingers is completed;
step S33: after the five fingers complete the grabbing of the grabbed object, the PVDF piezoelectric sensors 9 on the index finger 1, the ring finger 3 and the thumb 6 do not generate voltage signals, and in the grabbing process of the grabbed object under the action of the mechanical arm:
friction force is generated between the grasped object and five fingers, if the value of the maximum friction force is larger than or equal to the gravity of the grasped object, relative sliding does not occur between the five fingers and the grasped object, no voltage signal is generated by the PVDF piezoelectric sensors 9 on the index finger 1, the ring finger 3 and the thumb 6, the five fingers can stably grasp the grasped object, if the value of the maximum friction force is smaller than the gravity of the grasped object, relative sliding occurs between the five fingers and the grasped object, voltage signals are generated by the PVDF piezoelectric sensors 9 on the index finger 1, the ring finger 3 and the thumb 6, and the five fingers do not stably grasp the grasped object.
If no voltage signal is generated on the PVDF piezoelectric sensors on the index finger 1, the ring finger 3 and the thumb 6, which indicates that no sliding occurs between the five fingers and the grasped object, the stable grasping of the grasped object by the five fingers is completed;
if at least one of the PVDF piezoelectric sensors 9 on the index finger 1, the ring finger 3 and the thumb 6 is detected to have a voltage signal generated, the sliding between the five fingers and the grasped object is indicated, and then the step S34 is carried out;
step S34: voltage signals generated by the PVDF piezoelectric sensor 9 are amplified, converted and filtered by an AD620 filtering and collecting module, then the voltage signals are output and sent to an STM32 control panel, then the STM32 control panel sends the obtained voltage signals to an upper computer, the upper computer gradually updates and adjusts signal values corresponding to PWM duty cycle signal intervals corresponding to the hardness grade of a grasped object in an instruction program and sends the signal values to an STM32 control panel, the STM32 control panel controls three micro electric push rods to contract according to the instruction program which is gradually updated and adjusted each time, and then five fingers of a forefinger 1, a ring finger 3, a middle finger 2, a thumb 6 and a little finger 4 of the manipulator are bent;
step S35: repeating the steps S32-S34, and in step S34, if the adjusted PWM duty signal value reaches the maximum signal value corresponding to the PWM duty signal interval corresponding to the hardness level of the grasped object, it is still detected that the PVDF piezoelectric sensor 9 generates a voltage signal in the process of grasping the grasped object under the action of the robot arm in step S33, so as to prevent the grasped object from being damaged, directly abandon the grasping, and determine that the grasping fails, which indicates that the five fingers have not completed the stable grasping of the grasped object.
Fig. 7 is a voltage signal diagram obtained by the PVDF voltage sensor when the manipulator identifies objects with different hardness, fig. 7 is a voltage signal diagram obtained by the PVDF voltage sensor when the manipulator identifies the sponge, the Ecoflex, the PDMS, and the steel block in sequence from left to right, and voltage signals generated by the PVDF voltage sensor are also different when the manipulator identifies objects with different hardness.
According to the invention, the hard humanoid manipulator manufactured by 3D printing is used as a framework, so that soft and hard identification and self-adaptive stable grabbing of an object can be realized. However, since the flexible sensor is used to identify the physical properties of the object in the present invention, the present invention is not limited to the foregoing embodiments, and can be applied to various flexible robot systems.
Claims (6)
1. An object hardness identification and self-adaptive grabbing method based on a humanoid manipulator device is characterized in that: the method comprises the following specific steps:
step S1: setting the output force of the miniature electric push rod:
step S11: firstly, the upper computer sends a program instruction with a PWM duty ratio signal of 80 to an STM32 control board, the STM32 control board controls a first minitype electric push rod (16) to retract, further pulling the driving flexible rope (12) on the forefinger (1) to make the driving flexible rope (12) tensioned, bending the torsion spring (10) on the forefinger (1) to further make the forefinger (1) bent, bending the forefinger (1) to contact with the grasped object and finish knocking once, when the forefinger (1) finishes knocking, the PVDF piezoelectric sensor (9) on the index finger (1) generates a voltage signal, after the index finger (1) finishes knocking, the upper computer is an STM32 control board which sends a program instruction that the PWM duty ratio signal is-80, an STM32 control board controls a first minitype electric push rod (16) to extend out, further driving the driving flexible rope (12) on the forefinger (1) to be loosened, and the torsion spring (10) on the forefinger (1) deforms and restores and drives the forefinger (1) to restore to a straight state;
step S12: the voltage signal that PVDF piezoelectric sensor (9) on forefinger (1) produced is output after the module enlargies and the conversion filtering through filtering gathers and is sent to STM32 control panel, judges the voltage interval that voltage signal belongs to through STM32 control panel operation processing, then the hardness grade that the STM32 control panel obtained the voltage interval and corresponds obtains the hardness grade of being grabbed the object according to the voltage interval that voltage signal belongs to according to following corresponding relation, and will be grabbed the hardness grade of object and send for the host computer:
if the voltage interval in which the voltage signal is located is [ 5-10) mv, the hardness grade of the grasped object is A, if the voltage interval in which the voltage signal is located is [ 10-25) mv, the hardness grade of the grasped object is B, if the voltage interval in which the voltage signal is located is [ 25-60) mv, the hardness grade of the grasped object is C, and if the voltage interval in which the voltage signal is located is [ 60-150 ] mv, the hardness grade of the grasped object is D;
step S13: the upper computer judges a PWM duty ratio signal interval corresponding to the hardness grade of the grasped object and an output force interval corresponding to the PWM duty ratio signal interval according to the obtained hardness grade of the grasped object and the following corresponding relation:
if the hardness grade of the grasped object is A, a PWM duty cycle signal interval corresponding to the hardness grade A is [ 80-100 ], and an output force interval corresponding to the PWM duty cycle signal interval [ 80-100) is [ 2-2.5) N; if the hardness grade of the grasped object is B, the PWM duty cycle signal interval corresponding to the hardness grade B is [ 100-140 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 100-140 ] is [ 2.5-5) N; if the hardness grade of the grasped object is C, the PWM duty cycle signal interval corresponding to the hardness grade C is [ 140-190 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 140-190) is [ 5-8.5) N; if the hardness grade of the grasped object is D, the PWM duty cycle signal interval corresponding to the hardness grade D is [ 190-255 ], and the output force interval corresponding to the PWM duty cycle signal interval [ 190-255 ] is [ 8.5-12.5 ] N;
the upper computer sends a program instruction of a minimum signal value corresponding to a PWM duty cycle signal interval corresponding to the hardness grade of the grasped object to an STM32 control board according to the obtained hardness grade of the grasped object, and the STM32 control board controls the micro electric push rod to output an output force corresponding to the minimum signal value corresponding to the PWM duty cycle signal interval according to the obtained program instruction;
step S2: judging the hardness of the grasped object:
step S21: the STM32 control board controls the first miniature electric push by the program command of the minimum signal value corresponding to the PWM duty ratio signal section corresponding to the hardness grade obtained in step S13The rod (16) contracts, the driving flexible rope (12) on the forefinger (1) is tensioned, the torsion spring (10) on the forefinger (1) bends to further enable the forefinger (1) to bend until the forefinger (1) is not bent, the forefinger (1) contacts with a grasped object and finishes the pressing of the forefinger (1) on the grasped object, when the forefinger (1) finishes the pressing of the grasped object, the pressure F received by the forefinger (1) of the humanoid manipulator is F2Is equal to the output force corresponding to the minimum signal value corresponding to the PWM duty ratio signal interval output by the first miniature electric push rod (16);
step S22: in the bending process of the forefinger (1), the two liquid metal sensors (13) on the forefinger (1) are stretched, the two liquid metal sensors (13) output respective resistance signals, the forefinger (1) stops bending until the resistance values of the resistance signals output by the two liquid metal sensors (13) are not changed any more, the respective resistance signals output by the two liquid metal strain sensors (13) are sent to an STM32 control panel through a voltage division circuit, the STM32 control panel obtains two resistance values and sends the two resistance values to an upper computer, the upper computer converts the two resistance values into two corners through analysis and calculation, and then the upper computer further obtains fingertip displacement of the forefinger (1) through calculation according to the two corners;
step S23: the hardness sensitivity coefficient S is used as the measurement of hardness of the grasped object, and the displacement of the fingertip of the index finger is equal to the deformation x of the grasped object under pressure in the process that the index finger presses the grasped object2(ii) a The soft and hard sensitivity coefficient S is calculated using the following formula:
in the formula, K1Denotes the elastic coefficient, x, of the index finger tip2Representing the amount of compressive deformation of the gripped object, F2Indicating the pressure applied by the index finger of the manipulator when the pressing is finished;
obtaining the softness and hardness of the grasped object after obtaining the softness and hardness sensitivity coefficient S;
step S3: stably grabbing a grabbed object:
step S31: when the grabbed object is grabbed, the mechanical arm is fixed on the humanoid manipulator to realize that the grabbed object is grabbed after five fingers are contacted with the grabbed object; the STM32 control board controls the three micro electric push rods to contract according to the program instruction of the minimum signal value corresponding to the PWM duty ratio signal interval corresponding to the hardness grade obtained in the step S13, so that five fingers of the manipulator are bent;
step S32: when the five fingers bend to be in contact with a grasped object, voltage signals are generated by PVDF piezoelectric sensors (9) on the index finger (1), the ring finger (3) and the thumb (6), and when the pressure applied to the five fingers is equal to the output force corresponding to the minimum signal value corresponding to the PWM duty cycle signal interval output by the three miniature electric push rods, the grasping of the grasped object by the five fingers is completed;
step S33: after the five fingers complete the grabbing of the grabbed object, the PVDF piezoelectric sensors (9) on the index finger (1), the ring finger (3) and the thumb (6) do not generate voltage signals, and in the grabbing process of the grabbed object under the action of the mechanical arm:
if no voltage signal is generated by the PVDF piezoelectric sensors on the index finger (1), the ring finger (3) and the thumb (6), finishing stable grabbing of the grabbed object by five fingers;
if at least one of the PVDF piezoelectric sensors (9) on the index finger (1), the ring finger (3) and the thumb (6) is detected to have a voltage signal generated, sliding occurs between the five fingers and the grasped object, and then the step S34 is carried out;
step S34: voltage signals generated by the PVDF piezoelectric sensor (9) are amplified by the filtering acquisition module and are output and sent to the STM32 control panel after conversion and filtering, then the STM32 control panel sends the obtained voltage signals to the upper computer, the upper computer gradually updates and enlarges signal values corresponding to PWM duty cycle signal intervals corresponding to the hardness grade of a grasped object in a command program and sends the signal values to the STM32 control panel, and the STM32 control panel controls three miniature electric push rods to contract according to the command program which is gradually updated and enlarged each time, so that five fingers of the manipulator are bent;
step S35: and (3) repeating the steps S32-S34, and in the step S34, if the adjusted PWM duty ratio signal value reaches the maximum signal value corresponding to the PWM duty ratio signal interval corresponding to the hardness grade of the grasped object, detecting that a voltage signal is generated by the PVDF piezoelectric sensor (9) in the process of grasping the grasped object under the action of the mechanical arm in the step S33, preventing the grasped object from being damaged, directly abandoning the grasping, and judging the grasping failure.
2. The method for recognizing hardness and grabbing the object based on the humanoid manipulator device as claimed in claim 1, wherein:
the method adopts the following humanoid manipulator device which comprises a bracket (8), a humanoid manipulator, a sensing module, a control system and a driving module; the humanoid manipulator is fixed on the support (8), the sensing module is arranged on fingers of the humanoid manipulator, the driving module is connected to the back face of a palm of the humanoid manipulator, the input end of the control system is electrically connected with the sensing module, and the output end of the control system is electrically connected with the driving module.
3. The method for recognizing the hardness and the self-adaption grabbing of the object based on the humanoid manipulator device as claimed in claim 2, wherein the method comprises the following steps: the human-simulated manipulator comprises a palm (7) and five fingers connected with the palm (7), wherein the five fingers comprise an index finger (1), a middle finger (2), a ring finger (3), a thumb (6) and a little finger (4); the sensing module comprises a liquid metal flexible strain sensor (13) and a PVDF piezoelectric sensor (9); the control system comprises an STM32 control panel and an upper computer; the driving module comprises three micro electric push rods of a first micro electric push rod (16), a second micro electric push rod (17) and a third micro electric push rod (18), a torsion spring (10) and a driving flexible rope (12);
PVDF piezoelectric sensors (9) are pasted on fingertips of an index finger (1), a ring finger (3) and a thumb (6) of the humanoid manipulator, liquid metal flexible strain sensors (13) are pasted on a far knuckle rotating joint (11) and a middle knuckle rotating joint (15) of the index finger (1) of the humanoid manipulator, torsion springs (10) are respectively arranged at the far knuckle rotating joints of the index finger (1), the middle finger (2), the ring finger (3) and the little finger (4), torsion springs (10) are respectively arranged at the middle knuckle joints of the index finger (1), the middle finger (2), the ring finger (3) and the little finger (4), the torsion springs (10) are respectively arranged at the far knuckle rotating joint of the thumb (6), the torsion springs (10) are respectively arranged at five fingers of the humanoid manipulator, the near knuckle rotating joints of the five fingers of the humanoid manipulator are respectively vertically fixed on a palm (7), and no bending occurs; the three miniature electric push rods are vertically fixed on the back of the palm (7), one end of a driving flexible rope (12) is fixed on the far knuckle of each finger, and the other end of the driving flexible rope (12) sequentially penetrates through the far knuckle, the far knuckle rotating joint, the middle knuckle rotating joint and the near knuckle of each finger and then is tied to the miniature electric push rods on the back of the palm (7);
PVDF piezoelectric sensor (9) is connected an analog input end of STM32 control panel after the module amplification of collection of filtering and conversion filtering, and flexible strain transducer (13) of liquid metal is connected another input of STM32 control panel through dividing electric road roller electricity, and an output end electricity connection host computer of STM32 control panel, three miniature electric putter of another output electricity connection of STM32 control panel.
4. The method for recognizing the hardness and the self-adaption grabbing of the object based on the humanoid manipulator device as claimed in claim 2, wherein the method comprises the following steps: the upper computer and the STM32 control board control the three miniature electric push rods to contract, so that the driving flexible ropes (12) of each finger are tensioned, the torsion springs (10) at the knuckle rotating joints of each finger are bent, and the fingers are driven to bend; the upper computer and the STM32 control board are stretched by controlling the three miniature electric push rods, so that the driving flexible ropes (12) of each finger are loosened, and the extension of each finger is realized under the action of the torsion springs (10) at the rotary joints of each knuckle.
5. The method for recognizing the hardness and the self-adaption grabbing of the object based on the humanoid manipulator device as claimed in claim 2, wherein the method comprises the following steps: the drive flexible rope (12) on the forefinger (1) is tied on a first miniature electric push rod (16) close to the back of a palm (7) of the thumb, the drive flexible rope (12) on the middle finger (2), the ring finger (3) and the little finger (4) are tied on a second miniature electric push rod (17) in the middle of the back of the palm (7), and the drive flexible rope (12) on the little finger (6) is tied on a third miniature electric push rod (18) far away from the back of the palm (7) of the thumb.
6. The method for recognizing the hardness and the self-adaption grabbing of the object based on the humanoid manipulator device as claimed in claim 2, wherein the method comprises the following steps: the PVDF piezoelectric sensor (9) monitors dynamic force, when the force changes, the PVDF piezoelectric sensor (9) generates a voltage signal, and when the force does not change, the PVDF piezoelectric sensor (9) does not generate the voltage signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110199135.5A CN112976025B (en) | 2021-02-22 | 2021-02-22 | Object soft and hard recognition and self-adaptive grabbing method based on humanoid manipulator device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110199135.5A CN112976025B (en) | 2021-02-22 | 2021-02-22 | Object soft and hard recognition and self-adaptive grabbing method based on humanoid manipulator device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112976025A CN112976025A (en) | 2021-06-18 |
| CN112976025B true CN112976025B (en) | 2022-03-11 |
Family
ID=76350265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110199135.5A Active CN112976025B (en) | 2021-02-22 | 2021-02-22 | Object soft and hard recognition and self-adaptive grabbing method based on humanoid manipulator device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112976025B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111230911B (en) * | 2020-03-20 | 2021-05-04 | 燕山大学 | Flexible finger joint paw based on 4D printing and trajectory tracking control method thereof |
| CN113733149B (en) * | 2021-09-02 | 2024-03-19 | 浙江大学 | Soft finger manipulator for sensing fruit hardness and detection method |
| CN114193486A (en) * | 2021-11-18 | 2022-03-18 | 上海大学 | Pneumatic perception end effector of fruit and vegetable picking robot |
| CN114619445B (en) * | 2022-03-11 | 2024-05-03 | 北京理工大学 | Grabbing system |
| CN116766212B (en) * | 2023-08-17 | 2023-12-19 | 深圳市心流科技有限公司 | Bionic hand control method, bionic hand control device, bionic hand control equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4990815A (en) * | 1989-07-11 | 1991-02-05 | Lindner Douglas K | Robot gripper control system using PYDF piezoelectric sensors |
| CN106002995A (en) * | 2016-05-28 | 2016-10-12 | 上海大学 | Grabbing control system for five-finger anthropomorphic manipulator |
| CN107263470A (en) * | 2017-05-26 | 2017-10-20 | 吉林大学 | Soft grasping method for controlling robot based on multi-sensor information fusion |
| CN108789384A (en) * | 2018-09-03 | 2018-11-13 | 深圳市波心幻海科技有限公司 | A kind of flexible drive manipulator and the object identification method based on three-dimensional modeling |
| CN110039533A (en) * | 2019-04-17 | 2019-07-23 | 苏州柔性智能科技有限公司 | For detecting the multi-functional software manipulator of fruit maturity |
-
2021
- 2021-02-22 CN CN202110199135.5A patent/CN112976025B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4990815A (en) * | 1989-07-11 | 1991-02-05 | Lindner Douglas K | Robot gripper control system using PYDF piezoelectric sensors |
| CN106002995A (en) * | 2016-05-28 | 2016-10-12 | 上海大学 | Grabbing control system for five-finger anthropomorphic manipulator |
| CN107263470A (en) * | 2017-05-26 | 2017-10-20 | 吉林大学 | Soft grasping method for controlling robot based on multi-sensor information fusion |
| CN108789384A (en) * | 2018-09-03 | 2018-11-13 | 深圳市波心幻海科技有限公司 | A kind of flexible drive manipulator and the object identification method based on three-dimensional modeling |
| CN110039533A (en) * | 2019-04-17 | 2019-07-23 | 苏州柔性智能科技有限公司 | For detecting the multi-functional software manipulator of fruit maturity |
Non-Patent Citations (1)
| Title |
|---|
| "仿人机械手设计与硬度感知研究";张颖等;《东北大学学报（自然科学版）》;20200331;第41卷(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112976025A (en) | 2021-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112976025B (en) | Object soft and hard recognition and self-adaptive grabbing method based on humanoid manipulator device | |
| Carrozza et al. | Experimental analysis of an innovative prosthetic hand with proprioceptive sensors | |
| Ueda et al. | Development of the NAIST-Hand with vision-based tactile fingertip sensor | |
| Dong et al. | Stretchable human machine interface based on smart glove embedded with PDMS-CB strain sensors | |
| CN104665962B (en) | Wearable function strengthens machine hand system and its assisted finger and control method | |
| Tadano et al. | Development of grip amplified glove using bi-articular mechanism with pneumatic artificial rubber muscle | |
| Biddiss et al. | Electroactive polymeric sensors in hand prostheses: Bending response of an ionic polymer metal composite | |
| Yang et al. | A low-cost linkage-spring-tendon-integrated compliant anthropomorphic robotic hand: MCR-Hand III | |
| CN108453766B (en) | Multi-joint bionic dexterous artificial finger | |
| CN109199783B (en) | Control method for controlling stiffness of ankle joint rehabilitation equipment by using sEMG | |
| Wen et al. | Force-guided high-precision grasping control of fragile and deformable objects using semg-based force prediction | |
| Gerratt et al. | Stretchable capacitive tactile skin on humanoid robot fingers—First experiments and results | |
| CN113910279B (en) | Perception-driving integrated rigid-flexible coupling finger and flexible robot dexterous hand | |
| CN108673538B (en) | Mechanical finger and mechanical arm | |
| Controzzi et al. | Bioinspired fingertip for anthropomorphic robotic hands | |
| CN113146665B (en) | Dexterous hand with tactile feedback and variable rigidity and control method thereof | |
| Deng et al. | Design of a soft gripper with improved microfluidic tactile sensors for classification of deformable objects | |
| Van Volkinburg et al. | Development of a wearable controller for gesture-recognition-based applications using polyvinylidene fluoride | |
| JP4737099B2 (en) | Robot and robot control apparatus and control method | |
| Ang et al. | 3D printed soft pneumatic actuators with intent sensing for hand rehabilitative exoskeletons | |
| Feng et al. | A fiber-reinforced human-like soft robotic manipulator based on sEMG force estimation | |
| Song et al. | A Wearable Capacitive Friction Force Sensor for E-Skin | |
| CN214643717U (en) | A imitative people manipulator device that is used for object hardness to discern and self-adaptation snatchs | |
| Canbay et al. | Calibration of tactile/force sensors for grasping with the PRISMA Hand II | |
| Nojiri et al. | Sensing and control of friction mode for contact area variable surfaces (friction-variable surface structure) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210618 Assignee: Zhejiang cloud byte information technology Co.,Ltd. Assignor: JIANG University OF TECHNOLOGY Contract record no.: X2023980033010 Denomination of invention: Soft and hard object recognition and adaptive grasping method based on humanoid manipulator Granted publication date: 20220311 License type: Common License Record date: 20230228 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210618 Assignee: ZHEJIANG GUANGLI ENGINEERING MACHINERY Co.,Ltd. Assignor: JIANG University OF TECHNOLOGY Contract record no.: X2023980037308 Denomination of invention: Object Soft and Hard Recognition and Adaptive Grasping Method Based on Humanoid Robot Device Granted publication date: 20220311 License type: Common License Record date: 20230630 |