A two-dimensional multi-wave generator type harmonic reducer and adaptive industrial robot

Through the two-dimensional multi-head wave generator harmonic reducer with embedded memory alloy, adaptive overload protection and flexible deformability in extreme working conditions and non-structural environments are achieved, and the large size and inability to adapt to the environment in the prior art are solved. Changes in the problem, providing highly integrated and adaptive industrial robot joint solutions.

CN118149069BActive Publication Date: 2024-11-15ZHEJIANG INTERFACE TRANSMISSION CO LTD
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202410131326.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-29
Publication Date
2024-11-15
Estimated Expiration
2044-01-29

AI Technical Summary

Technical Problem

The existing harmonic reducer type electromechanical series joint cannot meet the application requirements of adaptive overload protection and adaptive flexibility and variable high-integrated adaptive industrial robots in extreme operating conditions and non-structural environments. It is large in size and heavy in weight. , and overload protection cannot be achieved under extreme operating conditions and cannot adapt to changes in non-structural environments.

Method used

The two-dimensional multi-head wave generator harmonic reducer with embedded memory alloy can realize the radial telescopic and peristalsis of the flexible wheel through the current driving of the shape memory alloy segment, forming a toothless transmission system to realize self-drive and self-sensing functions. The traditional motor drive structure is reduced, the volume is reduced, and the friction and wear resistance are improved through bionic microstructures to adapt to changes in different environments.

Benefits of technology

The harmonic reducer has been greatly reduced in size, meeting the needs of highly integrated adaptive robots, providing overload self-protection under extreme operating conditions and flexible deformability in non-structural environments, reducing transmission back-reduction and improving Transmission accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118149069B_ABST
    Figure CN118149069B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of industrial robots, and discloses a two-dimensional multi-head wave generator with embedded memory alloy, a harmonic reducer using the wave generator, and an industrial robot using the harmonic reducer. The wave generator includes a support portion, and a plurality of wave generating heads are radially arranged in the same plane along the circumferential direction of the support portion. The wave generating heads include a plurality of shape memory alloy segments, and each wave generating head is connected to a power supply. The power supply can pass a current with a phase difference to make the shape memory alloy segment stretch and creep, thereby changing the length of the wave generating head. The present invention can realize a significant reduction in the volume of the harmonic reducer, a reduction in weight, and flexible deformation and overload self-protection, and solves the problem that the existing harmonic reducer type electromechanical series integrated joint cannot meet the application requirements of highly integrated adaptive industrial robots with adaptive overload protection and adaptive flexibility under extreme working conditions and non-structural environments.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of industrial robots, and in particular to a two-dimensional multi-wave generator type harmonic reducer and a highly integrated adaptive industrial robot. Background Art

[0002] With the rapid development of industrial robots, the robot joint reducer and motor integrated system require a highly integrated high-power density integrated joint robot design with smaller and lighter volume. The harmonic gear reducer currently used in robot joints is widely used in the field of industrial robots due to its small size and light weight. However, the robot joint transmission using ordinary harmonic reducers is driven by the motor to drive the harmonic reducer input end based on the mechanical cam wave generator to excite. The motor and harmonic reducer are installed in series and integrated together. The integrated joint is large in size and weight, which can no longer meet the development needs of highly integrated industrial robots.

[0003] On the other hand, when the robot joint is subjected to extreme working conditions such as overload impact, the industrial robot joint based on the traditional harmonic reducer cannot achieve overrunning clutch and overload protection, so that the meshing gear of the robot joint harmonic reducer will be damaged or broken under the instantaneous super-strong impact, causing the robot to fail. Therefore, the transmission harmonic reducer type joint cannot meet the needs of industrial robots with adaptive overload self-protection under extreme working conditions.

[0004] Finally, for industrial robots operating in non-structural environments, since the working environment is changing at any time, it is not fixed, and it is complex and changeable, so the rigid non-deformable industrial robots using traditional harmonic reducers cannot meet the working conditions of non-structural environments, and the reducers they need also require variable transmission ratio movement and flexible volume. The main structure of the current harmonic reducer is composed of rigid wheels, flexible wheels, and wave generators, all of which are rigid components and cannot change their volume adaptively with environmental changes. Therefore, the transmission harmonic reducer type joint cannot meet the needs of adaptive flexible deformable industrial robots in non-structural environments.

[0005] In summary, the existing harmonic reducer type electromechanical series integrated joints cannot meet the application requirements of highly integrated adaptive industrial robots with adaptive overload protection and adaptive flexibility under extreme working conditions and non-structural environments. Summary of the invention

[0006] The present invention aims to provide a two-dimensional multi-wave generator type harmonic reducer with embedded memory alloy, so as to achieve a significant reduction in the volume of the harmonic reducer and solve the problem that the existing harmonic reducer is too large to meet the application requirements on some special micro robots.

[0007] To achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a two-dimensional multi-head wave generator, including a support part, on which a plurality of wave generating heads are radially arranged in the same plane along the circumferential direction, and the wave generating heads include a plurality of shape memory alloy segments, each of which is connected to a power supply, and the power supply can pass an electric current with a phase difference to make the shape memory alloy segment expand, contract and creep, thereby changing the length of the wave generating head.

[0008] Preferably, as an improvement, there are three wave generating heads.

[0009] Preferably, as an improvement, there are four wave generating heads.

[0010] Preferably, as an improvement, there are five wave generating heads.

[0011] Preferably, as an improvement, there are six wave generating heads.

[0012] Preferably, as an improvement, the wave generating head is made of a high damping material, and the shape memory alloy segment is embedded in the wave generating head.

[0013] Preferably, as an improvement, the wave generating head is made of shape memory alloy.

[0014] A two-dimensional multi-head wave generator type harmonic reducer comprises a rigid wheel, a flexible wheel and a wave generator. The rigid wheel is sleeved on the outside of the flexible wheel, and the flexible wheel is partially in friction contact with the rigid wheel. The wave generator adopts the above-mentioned two-dimensional multi-head wave generator with embedded memory alloy. By passing a current with a phase difference into the shape memory alloy segment of the wave generator, the shape memory alloy segment is stretched and creeped to deform, and the flexible wheel is driven to generate periodic radial stretching and creeping in the circumferential direction. A normal vector is formed between the flexible wheel and the rigid wheel in a two-dimensional plane. Then, the flexible wheel drives the rigid wheel to rotate in the circumferential direction through friction.

[0015] Preferably, as an improvement, the inner wall of the rigid wheel is an arc surface, the macroscopic surface texture of the wheel surface of the flexible wheel is a wedge-shaped texture that is continuous along the circumferential direction, the friction contact points between the wheel surface of the flexible wheel and the inner wall of the rigid wheel are multiple points, the circumference of the contact transmission interface between the rigid wheel and the flexible wheel is non-circular, and the contact transmission interface between the rigid wheel and the flexible wheel is a flexible deformable interface.

[0016] Preferably, as an improvement, the microscopic surface texture of the wheel surface of the flexible wheel is a plurality of suction cups, and the suction cups have a gradient elastic modulus decreasing from the surface to the inside in the radial direction of the flexible wheel.

[0017] Preferably, as an improvement, a current with a phase difference is passed through the shape memory alloy segment of the wave generator to deform the shape memory alloy segment, thereby driving the circumference of the flexible wheel to change, so that a variable transmission ratio transmission can be formed between the flexible wheel and the rigid wheel.

[0018] The principle and advantage of the present invention are: in practical application, the wave generator adopts a two-dimensional multi-head design, and the wave generator head of the wave generator adopts a shape memory alloy. The shape memory alloy (SMA) not only has self-damping and viscous damping characteristics, but also has high power density, large output force, flexibility, shape memory effect expansion and contraction and other functions. The SMA spring wire can be flexibly integrated in the multi-head wave generator of the two-dimensional plane, and the SAM is heated by current output to deform, and the multi-strand SMA spring wire integrated in the wave generator is controlled by the current phase difference to generate multi-wave and variable wave bionic creeping excitation, so that the wave generator can be self-driven and retracted in multiple directions, and can replace the motor to achieve self-drive; through the "resistance-length" change law of SMA, the displacement sensor is replaced, and the self-sensing and self-sensing functions similar to skeletal muscle spindles are exhibited. Therefore, integrating the SMA spring wire in the toothless transmission system can not only make the transmission system have self-driving and self-sensing characteristics, but also can use its superelasticity to overcome the rigidity and large volume of the traditional electromechanical integrated system, and realize flexible and small volume space layout and operation.

[0019] The transmission stiffness of the toothless contact interface is one of the factors affecting the load-bearing capacity. The transmission stiffness is affected by the size of the contact area. The change in the size of the contact area, through the self-driven and self-sensing deformation of the SMA material, changes in the number of radial support fluctuations of the regulating wave generator, which leads to changes in the contact area of ​​the toothless transmission pair, thereby causing changes in the contact transmission stiffness. The increase in the contact transmission area is conducive to achieving high transmission accuracy and low backlash. At the same time, due to the toothless transmission, overload adaptive protection can be achieved to prevent damage to the traditional reducer and breakage of the meshing tooth pair.

[0020] The use of embedded shape memory alloy wire and its power supply control replaces the process of the prior art motor-driven wave generator to drive the flexible wheel, and realizes the self-driven shape change of the wave generator. The wave generator of the present invention integrates the functions of the traditional wave generator and the motor, and does not affect the performance of the harmonic reducer. At the same time, the traditional motor drive structure is omitted, thereby achieving a significant reduction in the volume of the harmonic reducer. Compared with the conventional friction-type continuously variable transmission, the flexible wheel of the present invention can form a toothless transmission with the rigid wheel. The toothless transmission system has a simple structure, a small and compact volume, and the rigid wheel is not limited to the circular rigid structure of the prior art. The rigid wheel can adopt a variety of shapes. As long as it maintains multi-point friction transmission with the flexible wheel, the appearance shape of the rigid wheel can also be flexibly changed. In addition, the limitation of the application scenario is weakened, and it can be applied to special micro robots and other unconventional application scenarios.

[0021] The wheel surface of the flexible wheel adopts a composite bionic texture of an octopus sucker and a tree frog's fingertip at the microscopic level. The groove morphology on the surface of the octopus sucker helps to improve adhesion; the modulus of the tree frog's foot gradually decreases from the surface to the inside, and such a gradient modulus helps to improve adhesion and improve the friction and wear resistance of the structure. Inspired by the adhesion and friction structural characteristics of these two animals, the present invention combines the surface microstructure and performance of the octopus sucker and the tree frog's foot to perform coupled bionic design, and constructs a new type of gradient modulus composite bionic micro-suction cup texture, which improves its adhesion, friction, wear resistance and resistance to alternating fatigue failure, and can greatly enhance the friction coefficient in contact with the inner wall of the rigid wheel. The wheel surface adopts a wedge-shaped texture on a macro scale, and the toothless transmission contacts the working wedge-shaped texture interface. While being subjected to the wedge-shaped extrusion effect generated by the transmission on the inclined surface, it can enhance the wedge-shaped extrusion effect to increase the contact pressure, which helps to improve the contact friction of the transmission mechanism. After being subjected to the deformation force of the shape memory alloy, the extrusion pressure in contact with the inner wall of the rigid wheel can be enhanced, thereby greatly enhancing the contact friction between the flexible wheel and the rigid wheel, ensuring that the friction transmission between the flexible wheel and the rigid wheel is stable and reliable, and resisting alternating fatigue failure.

[0022] The harmonic reducer of this scheme is a flexible harmonic reducer. By using an embedded shape memory alloy on the wave generator, the strain of the shape memory alloy is controlled by powering on, so that the flexible wheel can directly telescope and creep in the radial section in situ, thereby realizing toothless friction transmission between the flexible wheel and the rigid wheel. The contact point between the flexible wheel and the rigid wheel is changed by the strain of the shape memory alloy under the phase difference current, so as to realize the friction transmission between the flexible wheel and the rigid wheel. The contact transmission interface between the flexible wheel and the rigid wheel can be circular or non-circular in shape. The contact interface is different from the traditional rigid non-deformable interface and can be a rigid interface or a flexible variable interface. Such a harmonic reducer eliminates the driving structure of the traditional wave generator, greatly reduces the overall volume, and the transmission ratio can be adjusted and changed, which can effectively meet the application requirements of highly integrated adaptive robots. It can also be used in unstructured application scenarios. For example, when the bionic snake moves and shuttles in an unstructured environment with different sizes of apertures underground, the flexible deformation of the joints can adapt to the sizes of different apertures, thereby realizing the application needs of adaptive industrial robots in unstructured environments; it can also be used when the robot joints are subjected to extreme working conditions such as overload impact. Since the toothless transmission cannot achieve overrunning clutch and overload protection compared to the toothed transmission of the traditional harmonic reducer, the meshing gears of the harmonic reducer of the robot joints will be damaged or broken under instantaneous super-strong impact, causing the robot to fail, thereby realizing the needs of industrial robots with adaptive overload self-protection under extreme working conditions.

[0023] A highly integrated adaptive industrial robot adopts the integrated electromechanical joint of the above-mentioned two-dimensional multi-head wave generator type harmonic reducer. Such an industrial robot joint is small in size and light in weight, and is highly integrated with electromechanical integration; the toothless transmission at the joint can achieve overload impact protection, prevent overload damage or destruction of the reducer, and achieve adaptive overload self-protection of the industrial robot under extreme working conditions; the joint structure can be flexibly deformed to adapt to the flexible and variable adaptation of the industrial robot operating in a non-structural complex and changeable environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a schematic diagram of the structure of Example 1 of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of Example 2 of the present invention.

[0026] Figure 3 This is a schematic diagram of the structure of Example 3 of the present invention.

[0027] Figure 4 This is a schematic diagram of the structure of Example 4 of the present invention.

[0028] Figure 5 This is a schematic diagram of the structure of Example 7 of the present invention.

[0029] Figure 6 Schematic diagram of the macro texture of the flexible wheel surface in Example 7 of the present invention.

[0030] Figure 7 Schematic diagram of the microscopic texture of the flexible wheel surface in Example 7 of the present invention. DETAILED DESCRIPTION

[0031] The following is further described in detail through specific implementation methods:

[0032] The reference numerals in the drawings of the specification include: a rigid wheel 1 , a wave generating head 2 , a flexible wheel 3 , and a supporting portion 4 .

[0033] Embodiment 1, basically as attached Figure 1 As shown: a two-dimensional multi-head wave generator with embedded shape memory alloy, including a support part 4, on which four wave generating heads 2 are radially arranged in the same plane along the circumferential direction, and the wave generating heads 2 are made of high damping material such as rubber alloy, and each wave generating head 2 is embedded with a plurality of shape memory alloy segments along the length direction, and the shape memory alloy segments are spring-shaped shape memory alloy wires, and each wave generating head 2 is connected to a power supply, and the power supply can pass a current with a phase difference to make the shape memory alloy segment expand and contract and creep, thereby changing the length of the wave generating head 2.

[0034] Embodiment 2, this embodiment differs from embodiment 1 only in that, Figure 2 As shown, there are three wave generating heads 2.

[0035] Embodiment 3, this embodiment differs from embodiment 1 only in that, Figure 3 As shown, there are five wave generating heads 2.

[0036] Embodiment 4, this embodiment differs from embodiment 1 only in that, Figure 4 As shown, there are six wave generating heads 2.

[0037] Embodiment 5: The difference between this embodiment and embodiment 1 is that the number of wave generating heads 2 is eight.

[0038] Embodiment 6: The difference between this embodiment and embodiment 1 is that the wave generating head 2 is made of shape memory alloy as a whole.

[0039] Embodiment 7, a two-dimensional multi-wave generator type harmonic reducer, such as Figure 5 As shown, it includes a rigid wheel 1, a flexible wheel 3 and a wave generator. The rigid wheel 1 is sleeved on the outside of the flexible wheel 3, and the inner wall of the rigid wheel 1 is a circular arc surface. Figure 6 As shown in FIG. 1 , the macroscopic surface texture of the wheel surface of the flexible wheel 3 is a wedge-shaped texture that is continuous along the circumferential direction, such as Figure 7 As shown, the microscopic surface texture of the wheel surface of the flexible wheel 3 is a plurality of suction cups, and the suction cups have a gradient elastic modulus decreasing from the surface to the inside in the radial direction of the flexible wheel 3. The flexible wheel 3 is partially in friction contact with the rigid wheel 1. The wave generator adopts a two-dimensional multi-head wave generator with embedded memory alloy in Example 1.

[0040] During the specific implementation process, a current with a phase difference is introduced into the shape memory alloy segment of the wave generator to cause the shape memory alloy segment to expand and contract and creep, so that the lengths of the multiple wave generating heads 2 of the wave generator change, and the wave generating heads 2 push the flexible wheel 3, driving the flexible wheel 3 to produce periodic radial expansion and contraction creep in the circumferential direction, so that the flexible wheel 3 is elliptical, and the two ends of its long axis are in friction contact with the inner wall of the steel wheel, and a normal vector is formed between the flexible wheel 3 and the rigid wheel 1. The surface micro-texture of the flexible wheel 3 is a structure of bionic octopus suction cup combined with the fingertips of tree frogs, which increases the contact friction coefficient with the inner wall of the rigid wheel 1, and the macro-texture of the surface of the flexible wheel 3 is a wedge-shaped structure. When the flexible wheel 3 and the rigid wheel 1 are performing inclined plane transmission, the wedge-shaped texture forms a wedge-shaped extrusion on the transmission interface, and the extrusion effect is enhanced to increase the contact pressure, so that the contact friction between the flexible wheel 3 and the rigid wheel 1 is increased, and then the flexible wheel 3 drives the rigid wheel 1 to rotate by friction in the circumferential direction.

[0041] Embodiment 8, a highly integrated adaptive industrial robot, uses a two-dimensional multi-head wave generator type harmonic reducer as in embodiment 7 as a transmission component at the joints between the robot arms.

[0042] The present invention also proposes a neural network PID control method based on a resistance self-sensing model. First, the resistance value of the shape memory alloy is sampled, and then converted into the calculated strain size after the self-sensing model, and then the output signal of the feedback link is subtracted from the reference input signal, and the error is input to the controller. The generated control signal is power amplified and input to the SMA to adjust the heating power of the driver to obtain the actual output strain, thereby realizing the position tracking control of the driver for the shape memory alloy (SMA), and finally generating periodic changes through the multi-head (flexible wheel) of the current control wave generator with phase difference.

[0043] In practical applications, using the change in SMA resistance as a feedback signal for the control system can reduce the cost of using sensors and avoid increasing the size and weight of the device. Before establishing a resistance self-sensing model, we must first study the basic resistance characteristics of SMA. Next, we analyze the resistivity of SMA. The SMA resistance can be expressed as:

[0044]

[0045] Where R represents the resistance of SMA, L and S are the length and cross-sectional area of ​​SMA respectively, and ρ represents the resistivity of SMA. The change of resistivity with temperature T can be expressed as:

[0046] ρ=ρ 0 (1+aT) (1.2)

[0047] Where 0 represents the resistivity at 0°C and a is the temperature coefficient of resistance.

[0048] SMA can be divided into three phases: martensite, austenite and R phase. They have different crystal structures and exhibit different electrical and mechanical properties. The R phase makes the resistance analysis more complicated. Therefore, for the convenience of analysis, it is assumed that the resistance characteristics of SMA are mainly determined by the volume fraction of martensite and austenite. Therefore, according to the Brinson constitutive model, the formula for calculating the resistivity of SMA is as follows:

[0049] ρ=ρ M +(ξ AT -ξ AS )(ρ A -ρ M ) (1.3)

[0050] Among them, ξ A and M are the volume fractions of austenite and martensite, respectively, and ρ A and ρ M are the resistivity of austenite and martensite, ξ AT and ASThe temperature affects the volume fraction of austenite and the stress affects the volume fraction of austenite. It can be seen that the resistance self-sensing model derived based on the formula has many parameters and is relatively complex. It is difficult to establish an accurate resistance self-sensing model from a mechanism perspective and needs to be determined through experiments. In the feedback control system, resistance is converted into strain information, and a self-sensing model based on the resistance-strain curve needs to be established. However, similar to the obvious hysteresis gap shown in the current-strain curve, in order to better establish the resistance self-sensing model of SMA, the method of applying prestress to SMA can be used to reduce the hysteresis loop width to improve the model accuracy. In order to quantitatively study the influence of prestress on the hysteresis loop width of the curve, the calculation formula of the hysteresis loop width is defined as follows:

[0051]

[0052] Among them, H gap is the hysteresis width, s hi and ci is the strain value of the heating and cooling paths, s max is the maximum strain value.

[0053] Fitting the data with a high-order polynomial helps improve the accuracy of self-sensing modeling. In order to facilitate the use of the self-sensing model in embedded programs, the resistance is normalized. The resistance normalization formula is as follows:

[0054]

[0055] Where R max , R min and R are the maximum resistance, minimum resistance, and actual resistance, respectively, and r is the normalized resistance calculation result.

[0056] A sixth-order polynomial is used to fit the stable data after 50 trainings, and the polynomial equation obtained by fitting is as follows:

[0057]

[0058] Where r is the resistance value and s(r) is the corresponding calculated strain.

[0059] The essence of the resistor self-sensing model is to estimate strain through resistance, and during position control, the calculated strain is used instead of the displacement signal measured by the actual sensor as the feedback signal of the control system.

[0060] The position control algorithm of the SMA actuator consists of two parts: the traditional proportional-integral-derivative (PID) controller and the BP neural network. The discrete form of the incremental PID is given by the following equation:

[0061]

[0062] Among them, e(k), kp (k), k i (k) and k d (k) are the error, proportional gain, integral gain and differential gain at time k respectively.

[0063] The PID parameters are automatically adjusted by the neural network according to the error feedback, and the expression is as follows:

[0064]

[0065] Compared with traditional PID, neural network PID is more suitable for the control of nonlinear objects due to its automatic parameter adjustment capability. The process of BP neural network is mainly divided into two stages: forward propagation and back propagation. Among them, forward propagation is the process of calculating the output of the network, and the input data is passed from the input layer to the output layer layer by layer. Each layer of neurons performs weighted summation of the input data, and then generates the next layer of input through the activation function. Back propagation is the process of optimizing the network weights. By calculating the error between the output layer and the actual target, the error is propagated forward layer by layer, and the weights are updated to reduce the error. These two processes are repeated until the network converges.

[0066] Forward propagation is mainly used to calculate the output of the network. The S function is used as the activation function. The calculation formula is as follows:

[0067]

[0068] The superscript (1) represents the input layer, and the superscript (2) represents the hidden layer. X i (k) are the input and output of the input layer respectively. are the input and output of the hidden layer, respectively. is the weight of the line from the i-th neuron in the input layer to the j-th neuron in the hidden layer at time k-1. There are m neurons in the hidden layer.

[0069] Back propagation is mainly used to adjust and update the weights between the layers of the neural network. The delta learning algorithm is used. At the same time, in order to increase the convergence speed, the inertia term is introduced to make the weight coefficient adjustment more stable. The calculation formula is as follows:

[0070]

[0071]

[0072] Wherein, α is the inertia coefficient, 0≤α<1.

[0073] The weights between layers are repeatedly corrected through the above steps to obtain real-time adjusted PID parameters for use in the position control algorithm.

[0074] In this method, the resistance self-sensing model plus neural network control technology is used to realize the closed-loop control of the entire flexible harmonic reducer. According to the resistance change characteristics of the SMA material, the resistance-strain curve is analyzed, and appropriate prestress is applied to improve the influence of the hysteresis loop width. Then, a resistance self-sensing model with a unified heating and cooling path of the SMA material is established by high-order polynomial fitting. The purpose of eliminating the position sensor is achieved by detecting the resistance as the feedback signal. Based on this resistance self-sensing model, the feedback control scheme is studied to form a closed-loop position control of the shape memory alloy, compensate for the hysteresis nonlinearity of the SMA, and show good control performance for the nonlinear system.

[0075] The above is only an embodiment of the present invention, and the common knowledge such as the known specific technical solutions and / or characteristics in the solution is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several modifications and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A two-dimensional multi-wave generator type harmonic reducer, comprising a rigid wheel, a flexible wheel and a wave generator, wherein the rigid wheel is sleeved outside the flexible wheel, and characterized in that: The flexible wheel is partially in frictional contact with the rigid wheel. The wave generator adopts a two-dimensional multi-head wave generator with embedded memory alloy. The two-dimensional multi-head wave generator includes a support part. A plurality of wave generating heads are radially arranged in the same plane along the circumferential direction of the support part. The wave generating heads include a plurality of shape memory alloy segments. Each wave generating head is connected to a power supply. The power supply can pass a current with a phase difference to make the shape memory alloy segment stretch and creep, thereby changing the length of the wave generating head. By passing a current with a phase difference into the shape memory alloy segment of the wave generator, the shape memory alloy segment The memristor alloy segment stretches and creeps to drive the flexible wheel to produce periodic radial stretching and creeping in the circumferential direction, and then the flexible wheel drives the rigid wheel to rotate through friction in the circumferential direction, and a curved surface motion of the normal vector in a two-dimensional plane is formed between the flexible wheel and the rigid wheel; the inner wall of the rigid wheel is an arc curved surface, the macroscopic surface texture of the wheel surface of the flexible wheel is a wedge-shaped texture continuous along the circumferential direction, the friction contact points between the wheel surface of the flexible wheel and the inner wall of the rigid wheel are multiple points, the circumference of the contact transmission interface between the rigid wheel and the flexible wheel is non-circular, and the contact transmission interface between the rigid wheel and the flexible wheel is a flexible deformable interface.

2. A two-dimensional multi-wave generator type harmonic reducer according to claim 1, characterized in that: There are three wave generating heads.

3. A two-dimensional multi-wave generator type harmonic reducer according to claim 1, characterized in that: There are four wave generating heads.

4. A two-dimensional multi-wave generator type harmonic reducer according to claim 1, characterized in that: There are five wave generating heads.

5. A two-dimensional multi-wave generator type harmonic reducer according to claim 1, characterized in that: There are six wave generating heads.

6. A two-dimensional multi-wave generator type harmonic reducer according to any one of claims 1 to 5, characterized in that: The microscopic surface texture of the wheel surface of the flexible wheel is a plurality of suction cups, and the suction cups have a gradient elastic modulus decreasing from the surface to the inside in the radial direction of the flexible wheel, thereby realizing friction transmission and overload adaptive protection at the same time.

7. A two-dimensional multi-wave generator type harmonic reducer according to claim 6, characterized in that: By passing a current with a phase difference into the shape memory alloy segment of the wave generator to deform the shape memory alloy segment, the circumference of the flexible wheel is driven to change, so that a variable transmission ratio transmission can be formed between the flexible wheel and the rigid wheel.

8. An adaptive industrial robot, characterized in that: A harmonic reducer according to any one of claims 1 to 7 is used to meet the requirements of adaptive overload protection under extreme working conditions and adaptive flexible and variable highly integrated adaptive industrial robots in non-structural environments.

Citation Information

Patent Citations

  • Method for reducing drive eccentric errors of harmonic gear and automatic aligning wave generator

    CN102705451A

  • Actuator

    CN104769236A

  • Strain wave gearing and wave generator

    CN107210686A

  • Harmonic speed reduction clutch integration device

    CN109058423A

  • Harmonic gear transmission device and pneumatic wave generator thereof

    CN203130930U