A method and device for detecting robot foot contact and sliding based on fuzzy logic

Abstract

This invention provides a method and device for detecting robot foot contact and sliding based on fuzzy logic, comprising: using a proprioceptive inertial measurement unit installed on the foot of a humanoid robot to collect the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot; performing Kalman filtering on the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot to obtain the linear acceleration, linear velocity, and angular velocity of the humanoid robot's foot in the world coordinate system; and using hierarchical fuzzy logic to detect the foot state of the humanoid robot. The first layer uses a generalized type 2 fuzzy C-means algorithm to calculate the probability of foot contact with the outside world based on the three-axis acceleration and three-axis angular velocity of the foot, and detects contact events that ensure the foot is standing. The second layer uses an interval type 2 fuzzy inference system to detect lateral sliding events of the foot based on the probability of foot contact with the outside world, the linear velocity of the foot, and the linear acceleration of the foot.

Description

Technical Field

[0001] This invention relates to the field of humanoid robot contact detection and motion control technology, and in particular to a robot foot contact and sliding detection method and device based on fuzzy logic. Background Technology

[0002] Humanoid robots, with their human-like locomotion characteristics, exhibit unparalleled mobility in unstructured environments compared to wheeled or tracked robots. However, their floating base dynamics and high center of gravity structure pose significant challenges to state estimation, particularly intermittent foot slippage and contact deformation, which severely impact motion stability and control accuracy. Existing contact detection solutions mainly fall into three categories: methods based on external sensors (such as plantar pressure sensors and force sensors) can directly acquire contact signals, but they can only measure normal contact and cannot capture lateral slippage. Furthermore, their reliability decreases in non-rigid terrains such as soft grass, while also increasing hardware complexity and cost. Dynamic methods based on proprioceptors avoid the limitations of external sensors, but the complex leg kinematics, limited support points, and metatarsal force characteristics of bipedal robots make it difficult to meet stability and accuracy requirements. Data-learning-based methods (such as convolutional neural networks and Gaussian mixture models) can adapt to unstructured terrain, but they rely on large amounts of training data or foot force sensors, and the black-box nature of deep learning models leads to insufficient interpretability and limited generalization ability. Furthermore, related technologies generally struggle to effectively handle the multidimensional uncertainties caused by sensor noise, terrain diversity, and the instantaneous nature of contact sliding. There is an urgent need for a contact detection method that requires no external sensors, is computationally efficient, and possesses strong robustness and generalization capabilities, providing reliable support for the dynamic motion control and state estimation of bipedal robots in complex environments. Summary of the Invention

[0003] To address the aforementioned problems in the existing technology, embodiments of the present invention provide a method and device for detecting robot foot contact and sliding based on fuzzy logic.

[0004] In a first aspect, embodiments of the present invention provide a robot foot contact and sliding detection method based on fuzzy logic, comprising: using a proprioceptive inertial measurement unit (IMU) installed on the foot of a humanoid robot to acquire the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot; performing Kalman filtering on the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot to obtain the linear acceleration, linear velocity, and angular velocity of the humanoid robot's foot in the world coordinate system; and using hierarchical fuzzy logic to detect the foot state of the humanoid robot. The first layer uses a generalized type 2 fuzzy C-means algorithm (GT2FCM) to calculate the probability of foot contact with the outside world based on the three-axis acceleration and three-axis angular velocity of the foot, and detects contact events that ensure the foot is standing. The second layer uses an interval type 2 fuzzy inference system (IT2FIS) to detect lateral sliding events of the foot based on the probability of foot contact with the outside world, the linear velocity of the foot, and the linear acceleration of the foot.

[0005] Based on the above method embodiments, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention uses a proprioceptive inertial measurement unit (IMU) installed on the foot of the humanoid robot to collect the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot. This includes: collecting data at a frequency of 1000 Hz, wherein the data contains Gaussian white noise, wherein the accelerometer Gaussian white noise is 0.01 m / s² to 0.05 m / s², and the gyroscope random walk angular frequency is 0.0001 radians per second per square root hertz to 0.001 radians per second per square root hertz.

[0006] Based on the above method embodiments, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention uses the generalized type 2 fuzzy C-means algorithm GT2FCM in the first layer to calculate the probability of foot contact with the outside world based on the foot's triaxial acceleration and triaxial angular velocity. This includes: classifying input and output data using a subtractive clustering algorithm to determine cluster centers and their number r; constructing a fuzzy rule matrix R and a second-level membership matrix SM, where the fuzzy rule matrix R contains r fuzzy rules, each corresponding to the uncertainty value of d-dimensional input and output variables, and the second-level membership matrix SM represents the weights of each uncertainty dimension; calculating the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM; using the generalized type 2 fuzzy C-means algorithm GT2FCM to calculate the membership degree between the input data and each fuzzy rule; and normalizing the membership degree to obtain the contact probability PO.

[0007] Based on the above method embodiments, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention includes the following step: calculating the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM.

[0008]

[0009]

[0010] Among them, UNR ij Let SM be the weighted representative value of the j-th variable in fuzzy rule i under uncertainty condition g, obtained from the second-level membership matrix SM. im R is the element in the i-th row and m-th column of the SM matrix; ijm Let unx be the i-th row and j-th column of the m-th two-dimensional matrix in the R three-dimensional matrix; r(d-1) Input element for uncertainty; uny r Output the r-th rule for uncertainty type reduction; x r(d-1)g y is the input element of the training set; rg Output elements for the training set.

[0011] Based on the above method embodiments, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention uses an interval type 2 fuzzy inference system (IT2FIS) in the second layer. It detects lateral sliding events of the foot based on the contact probability between the foot and the outside world, the linear velocity of the foot, and the linear acceleration of the foot. The input data includes the contact probability PO, the x-axis foot velocity and foot acceleration, and the y-axis foot velocity and foot acceleration. The time window of the interval type 2 fuzzy inference system (IT2FIS) is matched with the robot's running speed; the time window decreases as the speed increases. Fuzzy inference distinguishes between contact events that ensure the foot remains upright and lateral sliding events.

[0012] Based on the above method embodiments, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention further includes, in addition to, calculating the probability of foot contact with the outside world based on the foot's triaxial acceleration and triaxial angular velocity:

[0013] fv z =0

[0014] fa z =0

[0015]

[0016] Among them, v z a is the foot velocity along the z-axis; z The acceleration at the foot along the z-axis; Let f be the rate of change of foot acceleration along the z-axis; f is the robot foot; based on the foot velocity and acceleration along the x-axis, foot velocity and acceleration along the y-axis, and the three-axis angular velocity of the foot, the probability of foot contact with the outside world is generated using the generalized type 2 fuzzy C-means algorithm GT2FCM.

[0017] Based on the above method embodiments, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention further includes, after detecting the lateral sliding event of the foot, the following steps: using the multi-joint dynamics and contact MuJoCo simulation environment to simulate and verify the open-source humanoid robot. The verification set includes walking backwards on flat ground, walking with a predetermined stride, walking on unstructured ground, and walking up stairs, covering both structured and unstructured terrain.

[0018] Secondly, embodiments of the present invention provide a robot foot contact and sliding detection device based on fuzzy logic, comprising: a first main module for acquiring the triaxial acceleration and triaxial angular velocity of the humanoid robot's foot using a proprioceptive inertial measurement unit (IMU) installed on the foot of the humanoid robot; a second main module for performing Kalman filtering on the triaxial acceleration and triaxial angular velocity of the humanoid robot's foot to obtain the linear acceleration, linear velocity, and angular velocity of the humanoid robot's foot in the world coordinate system; a third main module for detecting the foot state of the humanoid robot using hierarchical fuzzy logic, wherein the first layer uses a generalized type 2 fuzzy C-means algorithm (GT2FCM) to calculate the probability of foot contact with the outside world based on the triaxial acceleration and triaxial angular velocity of the foot, and detects contact events that ensure the foot is standing; and a fourth main module for detecting lateral sliding events of the foot based on the interval type 2 fuzzy inference system (IT2FIS) used in the second layer.

[0019] Thirdly, embodiments of the present invention provide an electronic device, comprising:

[0020] At least one processor, at least one memory, and a communication interface; wherein,

[0021] The processor, memory, and communication interface communicate with each other;

[0022] The memory stores program instructions that can be executed by the processor. The processor calls the program instructions to execute the fuzzy logic-based robot foot contact and sliding detection method provided by any of the various implementations of the first aspect.

[0023] Fourthly, embodiments of the present invention provide a non-transitory computer-readable storage medium storing computer instructions that cause a computer to execute a fuzzy logic-based robot foot contact and sliding detection method provided by any of the various implementations of the first aspect.

[0024] The robot foot contact and sliding detection method and device based on fuzzy logic provided in this invention collect data through a foot proprioception inertial measurement unit (IMU), eliminating the need for external contact or force sensors, thus reducing hardware complexity and cost. It is suitable for the complex leg dynamics and model uncertainties of humanoid robots. The hierarchical fuzzy inference framework integrates general type 2 fuzzy clustering and interval type 2 fuzzy logic, effectively handling multi-dimensional uncertainties such as sensor noise and terrain diversity, and achieving real-time detection of contact and sliding. The contact detection accuracy exceeds 95% under various terrains and gait conditions, and its generalization ability is superior to deep learning methods, providing reliable support for stable movement and accurate state estimation of bipedal robots in complex environments. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 A schematic diagram of the robot foot contact and sliding detection method based on fuzzy logic provided in an embodiment of the present invention;

[0027] Figure 2 A schematic diagram of the structure of a robot foot contact and sliding detection device based on fuzzy logic provided in an embodiment of the present invention;

[0028] Figure 3 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention;

[0029] Figure 4 A schematic diagram of the prediction curve effect under four different environmental conditions provided in the embodiments of the present invention;

[0030] Figure 5 This is a schematic diagram illustrating the input and output effects of the sliding detection model provided in an embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In addition, the technical features of the various embodiments or individual embodiments provided by the present invention can be arbitrarily combined to form feasible technical solutions. Such combinations are not constrained by the order of steps and / or structural composition patterns, but must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention. If there are step numbers in the following embodiments, they are only set for ease of explanation and the order between steps is not limited. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0032] This invention provides a fuzzy logic-based method for detecting robot foot contact and sliding. (See also...) Figure 1 The method includes: using a proprioceptive inertial measurement unit (IMU) installed on the foot of a humanoid robot to collect the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot; performing Kalman filtering on the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot to obtain the linear acceleration, linear velocity, and angular velocity of the humanoid robot's foot in the world coordinate system; and using hierarchical fuzzy logic to detect the foot state of the humanoid robot. The first layer uses the generalized type 2 fuzzy C-means algorithm (GT2FCM) to calculate the probability of foot contact with the outside world based on the three-axis acceleration and three-axis angular velocity of the foot, and detects contact events that ensure the foot is standing. The second layer uses the interval type 2 fuzzy inference system (IT2FIS) to detect lateral sliding events of the foot based on the probability of foot contact with the outside world, the linear velocity of the foot, and the linear acceleration of the foot.

[0033] Based on the above method embodiments, as an optional embodiment, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention uses a proprioceptive inertial measurement unit (IMU) installed on the foot of the humanoid robot to collect the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot. This includes: collecting data at a frequency of 1000 Hz, wherein the data contains Gaussian white noise, wherein the accelerometer Gaussian white noise is 0.01 m / s² to 0.05 m / s², and the gyroscope random walk angular frequency is 0.0001 radians per second per square root hertz to 0.001 radians per second per square root hertz.

[0034] Sensor Configuration: Compared to information provided by proprioceptive inertial measurement units (IMUs), directly mounting an IMU on the foot provides a more direct reflection of foot posture. This helps in observing whether the foot is in contact with the ground from a kinematic perspective. While pressure sensors can visually reflect the force on the sole of the foot, they often fail to accurately reflect non-rigid contact and are prone to damage. Therefore, using external contact force sensors to make judgments from a kinematic rather than a dynamic perspective is more intuitive and reliable. Related technologies propose phase correction estimation for landing probability in quadrupedal landing planning; the pitch of the knee and hip joints can reflect the actual leg phase information, which is more reliable and intuitive than the expected phase.

[0035] Based on the above method embodiments, as an optional embodiment, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention uses the generalized type 2 fuzzy C-means algorithm GT2FCM in the first layer to calculate the probability of foot contact with the outside world based on the foot's triaxial acceleration and triaxial angular velocity. This includes: classifying input and output data using a subtractive clustering algorithm to determine cluster centers and their number r; constructing a fuzzy rule matrix R and a second-level membership matrix SM, where the fuzzy rule matrix R contains r fuzzy rules, each corresponding to the uncertainty value of d-dimensional input and output variables, and the second-level membership matrix SM represents the weights of each uncertainty dimension; calculating the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM; using the generalized type 2 fuzzy C-means algorithm GT2FCM to calculate the membership degree between the input data and each fuzzy rule; and normalizing the membership degree to obtain the contact probability PO.

[0036] Based on the above method embodiments, as an optional embodiment, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention includes the following step: calculating the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM.

[0037]

[0038]

[0039] Among them, UNR ij Let SM be the weighted representative value of the j-th variable in fuzzy rule i under uncertainty condition g, obtained from the second-level membership matrix SM. im R is the element in the i-th row and m-th column of the SM matrix; ijm Let unx be the i-th row and j-th column of the m-th two-dimensional matrix in the R three-dimensional matrix; r(d-1) Input element for uncertainty; uny r Output the r-th rule for uncertainty type reduction; x r(d-1)gy is the input element of the training set; rg Output elements for the training set.

[0040] Multidimensional uncertainty is a common phenomenon in bipedal locomotion. The inherent variability of sensor noise, the complexity of motion structures, the diversity of terrain features, and the instantaneous nature of contact sliding all contribute to the ambiguity of feature interpretation. Compared to interval type-2 and type-1 fuzzy systems, general type-2 fuzzy systems are better at managing the uncertainty of inputs and rules. Compared to interval type-2 fuzzy systems, the second membership degree of general type-2 fuzzy systems reflects the probability density of noise and rule distributions. General type-2 fuzzy systems are chosen as the contact prediction model. Traditional general type-2 fuzzy systems require a large number of iterative calculations due to their reduced construction complexity and types. Therefore, a rule-based general type-2 fuzzy C-Means algorithm is proposed. This method is highly effective in generalization and robustness to noise and high-dimensional datasets, and has been shown to preserve the uncertainty of type-2 fuzzy systems while compressing complexity, thus promoting system expression and iterative optimization. Adaptive GT2FCM further extends the application of GT2FCM to multi-input single-output systems to achieve the goal of clustering multidimensional variables. GT2FCM is a data-driven modeling framework capable of learning and storing information in real time without offline training. The algorithm is based on the three-dimensional rule R, which is expressed as formula (4). The second-level member matrix SM is represented by formula (3). The reliability of the quantification rule under different conditions is determined. According to the subtractive clustering algorithm, the number of clusters r is obtained by performing high-dimensional classification on the input and output data. The subtractive clustering algorithm is a density-based unsupervised clustering algorithm, the core of which is an iterative algorithm that treats each data point as a potential center. The algorithm selects high-density points as cluster centers and deletes neighboring points until all data are covered. Then, the density of data within the cluster is determined according to the ratio of the distance of data within the cluster to the range of the dataset, thereby determining the magnitude of uncertainty g. The data within the cluster are evenly distributed according to the uncertainty, and the uncertainty weights from the center to the edge are reduced sequentially to obtain the SM matrix.

[0041] The matrix representing the rd dimension of the rule-based type restoration uncertainty is calculated as shown in equations (1) and (2). Using the type restoration matrix UNR, the input membership degrees are calculated using the FCM algorithm as follows:

[0042]

[0043] Where UNRin is the type-restored input matrix; I n Dsti is the nth input signal; I is the nth input signal. n The Eulerian distance vector between the input vector of the i-th rule in UNRin and the input vector of the i-th rule; U i For the nth input signal I nMembership relationship with the i-th rule; m is a constant (can be 2); μ i C is the i-th intermediate coefficient; i normalized μ i Value; max indicates the maximum value; min indicates the minimum value. (Using C++) i Weighting the UNRout output matrix includes:

[0044]

[0045] Wherein, UNRout is the reduced matrix of the UNR output.

[0046] Based on the above method embodiments, as an optional embodiment, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention uses an interval type 2 fuzzy inference system (IT2FIS) in the second layer. Based on the foot's contact probability with the outside world, the foot's linear velocity, and the foot's linear acceleration, it detects lateral sliding events of the foot. This includes: input data including the contact probability PO, the x-axis foot velocity and foot acceleration, and the y-axis foot velocity and foot acceleration; the time window of the interval type 2 fuzzy inference system (IT2FIS) is matched with the robot's running speed, and the time window decreases as the speed increases. Fuzzy inference distinguishes between contact events that ensure foot standing and lateral sliding events of the foot.

[0047] Based on the above method embodiments, as an optional embodiment, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention further includes, in addition to, calculating the probability of foot contact with the outside world based on the foot's triaxial acceleration and triaxial angular velocity:

[0048] fv z =0(5)

[0049] fa z =0(6)

[0050]

[0051] Among them, v z a is the foot velocity along the z-axis; z The acceleration at the foot along the z-axis; Let f be the rate of change of foot acceleration along the z-axis; f is the robot foot; based on the foot velocity and acceleration along the x-axis, foot velocity and acceleration along the y-axis, and the three-axis angular velocity of the foot, the probability of foot contact with the outside world is generated using the generalized type 2 fuzzy C-means algorithm GT2FCM.

[0052] According to the kinematic analysis results, when the object is in stable contact, the three-axis velocities, three-axis angular velocities, and three-axis accelerations are all 0. When detecting foot contact with the ground, it is only necessary to consider the velocity and acceleration in the Z-axis direction under the world coordinate system to determine whether the foot has collided with the ground. Subsequently, in the frequency domain analysis of the collision attributes, it should be combined with the acceleration changes to distinguish them. In order to ensure rigid and stable contact with the environment, equations (5) to (7) must be satisfied.

[0053] Based on the above method embodiments, as an optional embodiment, the robot foot contact and sliding detection method based on fuzzy logic provided in this embodiment of the invention further includes, after detecting the lateral sliding event of the foot, the following steps: using the multi-joint dynamics and contact MuJoCo simulation environment to simulate and verify the open-source humanoid robot. The verification set includes walking backwards on flat ground, walking with a predetermined stride, walking on unstructured ground, and walking up stairs, covering both structured and unstructured terrain.

[0054] The fuzzy logic-based robot foot contact and sliding detection method provided in this invention collects data through a foot proprioceptive inertial measurement unit (IMU), eliminating the need for external contact or force sensors, thus reducing hardware complexity and cost. It is adaptable to the complex leg dynamics and model uncertainties of humanoid robots. The hierarchical fuzzy inference framework integrates general type-2 fuzzy clustering and interval type-2 fuzzy logic, effectively handling multi-dimensional uncertainties such as sensor noise and terrain diversity, achieving real-time detection of contact and sliding. The contact detection accuracy exceeds 95% under various terrains and gait conditions, and its generalization ability surpasses that of deep learning methods, providing reliable support for stable movement and accurate state estimation of bipedal robots in complex environments.

[0055] like Figure 4 As shown, the prediction curves under four different environmental conditions illustrate the performance of the GT2FCM model. The red curve represents the predicted probability, the blue curve represents the threshold-based contact state estimation, and the green curve corresponds to the actual ground contact state measured by the plantar sensor. Experimental results show that the accuracy of the proposed model exceeds 95% under various environments and gait conditions. Specifically, in a stairwell environment, the model robustly detects instantaneous contact events and step-down misalignment, ensuring stable performance even in intermittent foot-ground non-interaction. Figure 5 As shown, the short-distance sliding simulation used a contact probability model, which included the foot's displacement and velocity in the horizontal direction, as well as the angular velocity used for steady-state contact estimation. The results showed that the predicted probability of steady-state contact was significantly reduced; despite initial contact with the ground and vertical contact upon lifting, the foot was still identified as being in unsteady contact.

[0056] The proposed fuzzy inference model has been shown to achieve high detection accuracy (over 95% under various conditions) and a large area under the ROC curve. These results demonstrate the model's potential for reliable contact classification under different terrains and gaits without retraining, while also leveraging the potential provided by dynamic thresholds. Compared to the complex decision-making processes inherent in deep learning classifiers, fuzzy logic provides transparent decision rules, aiding in understanding the system's behavior. The model is capable of reasoning about contact and slippage as discrete entities, which is particularly advantageous in cases involving intermittent contact or minimal foot slippage. Despite the numerous advantages of the fuzzy inference framework, its current limitations must be acknowledged. Specifically, manually tuning the membership functions and rules of the slippage detection model is very time-consuming and often requires expertise to adapt to new robots or conditions.

[0057] The various embodiments of this invention are implemented through programmed processing using a device with processor functionality. Therefore, in practical engineering, the technical solutions and functions of the various embodiments of this invention can be encapsulated into various modules. Based on this reality, and building upon the above embodiments, this invention provides a fuzzy logic-based robot foot contact and sliding detection device, which is used to execute the fuzzy logic-based robot foot contact and sliding detection method in the above method embodiments. See also... Figure 2 The device includes: a first main module for acquiring the triaxial acceleration and triaxial angular velocity of the humanoid robot's foot using a proprioceptive inertial measurement unit (IMU) installed on the robot's foot; a second main module for performing Kalman filtering on the triaxial acceleration and angular velocity of the humanoid robot's foot to obtain the linear acceleration, linear velocity, and angular velocity of the humanoid robot's foot in the world coordinate system; a third main module for detecting the foot state of the humanoid robot using hierarchical fuzzy logic, wherein the first layer uses a generalized type 2 fuzzy C-means algorithm (GT2FCM) to calculate the probability of foot contact with the outside world based on the triaxial acceleration and angular velocity of the foot, and detects contact events that ensure the foot is standing; and a fourth main module for implementing a second layer using an interval type 2 fuzzy inference system (IT2FIS) to detect lateral sliding events of the foot based on the probability of foot contact with the outside world, the linear velocity of the foot, and the linear acceleration of the foot.

[0058] The robot foot contact and sliding detection device based on fuzzy logic provided in this embodiment of the invention adopts... Figure 2Several modules in the system collect data through foot proprioception inertial measurement units (IMUs), eliminating the need for external contact or force sensors, thus reducing hardware complexity and cost. This adapts to the complex leg dynamics and model uncertainties of humanoid robots. The hierarchical fuzzy inference framework integrates general type 2 fuzzy clustering and interval type 2 fuzzy logic, effectively handling multi-dimensional uncertainties such as sensor noise and terrain diversity, and achieving real-time detection of contact and sliding. The contact detection accuracy exceeds 95% under various terrains and gait conditions, and its generalization ability is superior to deep learning methods, providing reliable support for stable movement and accurate state estimation of bipedal robots in complex environments.

[0059] It should be noted that the apparatus in the device embodiments provided by the present invention can be used not only to implement the methods in the above method embodiments, but also to implement the methods in other method embodiments provided by the present invention. The difference lies only in the setting of corresponding functional modules. Its principle is basically the same as that of the above device embodiments provided by the present invention. As long as those skilled in the art, based on the above device embodiments and referring to the specific technical solutions in other method embodiments, obtain corresponding technical means and technical solutions composed of these technical means by combining technical features, and improve the apparatus in the above device embodiments while ensuring the practicality of the technical solutions, they can obtain corresponding device-type embodiments for implementing the methods in other method-type embodiments. For example:

[0060] Based on the above-described device embodiments, as an optional embodiment, the robot foot contact and sliding detection device based on fuzzy logic provided in this embodiment of the invention further includes: a first sub-module, used to implement the proprioceptive inertial measurement unit (IMU) installed on the foot of the humanoid robot to collect the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot, including: collecting data at a frequency of 1000 Hz, wherein the data contains Gaussian white noise, wherein the accelerometer Gaussian white noise is 0.01 m / s² to 0.05 m / s², and the gyroscope random walk angular frequency is 0.0001 radians per second per square root hertz to 0.001 radians per second per square root hertz.

[0061] Based on the above-described device embodiments, as an optional embodiment, the robot foot contact and sliding detection device based on fuzzy logic provided in this embodiment of the invention further includes: a second submodule, used to implement the first layer using the generalized type 2 fuzzy C-means algorithm GT2FCM to calculate the probability of foot contact with the outside world based on the foot's triaxial acceleration and triaxial angular velocity, including: classifying the input and output data using a subtractive clustering algorithm to determine the cluster centers and their number r; constructing a fuzzy rule matrix R and a second-level membership matrix SM, wherein the fuzzy rule matrix R contains r fuzzy rules, each fuzzy rule corresponds to the uncertainty value of the d-dimensional input and output variables, and the second-level membership matrix SM is the weight of each uncertainty dimension; calculating the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM, using the generalized type 2 fuzzy C-means algorithm GT2FCM to calculate the membership degree between the input data and each fuzzy rule, and obtaining the contact probability PO after normalizing the membership degree.

[0062] Based on the above-described device embodiments, as an optional embodiment, the robot foot contact and sliding detection device based on fuzzy logic provided in this embodiment of the invention further includes: a third submodule, used to implement the calculation of the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM, including:

[0063]

[0064] Among them, UNR ij Let SM be the weighted representative value of the j-th variable in fuzzy rule i under uncertainty condition g, obtained from the second-level membership matrix SM. im R is the element in the i-th row and m-th column of the SM matrix; ijm Let unx be the i-th row and j-th column of the m-th two-dimensional matrix in the R three-dimensional matrix; r(d-1) Input element for uncertainty; uny r Output the r-th rule for uncertainty type reduction; x r(d-1)g y is the input element of the training set; rg Output elements for the training set.

[0065] Based on the above-described device embodiments, as an optional embodiment, the robot foot contact and sliding detection device based on fuzzy logic provided in this embodiment of the invention further includes: a fourth submodule, used to implement the second layer using an interval type 2 fuzzy inference system (IT2FIS) to detect lateral sliding events of the foot based on the contact probability between the foot and the outside world, the linear velocity of the foot, and the linear acceleration of the foot, including: input data including the contact probability PO, the x-axis foot velocity and foot acceleration, and the y-axis foot velocity and foot acceleration; the time window of the interval type 2 fuzzy inference system (IT2FIS) is matched with the robot's running speed, and the time window decreases as the speed increases, and fuzzy inference is used to distinguish between contact events that ensure the foot is standing and lateral sliding events of the foot.

[0066] Based on the above-described device embodiments, as an optional embodiment, the robot foot contact and sliding detection device based on fuzzy logic provided in this embodiment of the invention further includes: a fifth submodule, used to calculate the probability of foot contact with the outside world based on the foot's triaxial acceleration and triaxial angular velocity, and further includes:

[0067] fv z =0

[0068] fa z =0

[0069]

[0070] Among them, v z a is the foot velocity along the z-axis; z The acceleration at the foot along the z-axis; Let f be the rate of change of foot acceleration along the z-axis; f is the robot foot; based on the foot velocity and acceleration along the x-axis, foot velocity and acceleration along the y-axis, and the three-axis angular velocity of the foot, the probability of foot contact with the outside world is generated using the generalized type 2 fuzzy C-means algorithm GT2FCM.

[0071] Based on the above device embodiments, as an optional embodiment, the robot foot contact and sliding detection device based on fuzzy logic provided in this embodiment of the invention further includes: a sixth submodule, used to detect the lateral sliding event of the foot, and further includes: using a multi-joint dynamics and contact MuJoCo simulation environment to simulate and verify the open-source humanoid robot, the verification set including walking backwards on flat ground, walking with a predetermined stride, walking on unstructured ground and walking up stairs, covering structured and unstructured terrain.

[0072] The method in this embodiment of the invention is implemented using an electronic device; therefore, it is necessary to introduce the relevant electronic device. For this purpose, this embodiment of the invention provides an electronic device, such as... Figure 3As shown, the electronic device includes at least one processor, a communications interface, at least one memory, and a communications bus, wherein the at least one processor, the communications interface, and the at least one memory communicate with each other via the communications bus. The at least one processor can invoke logical instructions stored in the at least one memory to execute all or part of the steps of the methods provided in the foregoing method embodiments.

[0073] Furthermore, when the logical instructions in at least one of the aforementioned memories can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various method embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0074] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0075] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0076] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Based on this understanding, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or sometimes in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0077] It should be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Any expressions such as "predetermined threshold," "preset threshold," etc., without specifying a particular value, can be determined by those skilled in the art through simple experimentation or appropriate adjustments.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A robot foot contact and sliding detection method based on fuzzy logic, characterized in that, include: A proprioceptive inertial measurement unit (IMU) installed on the foot of the humanoid robot was used to collect the three-axis acceleration and three-axis angular velocity of the foot of the humanoid robot. Kalman filtering is applied to the triaxial acceleration and angular velocity of the humanoid robot's feet to obtain the linear acceleration, linear velocity, and angular velocity of the feet in the world coordinate system. Hierarchical fuzzy logic is used to detect the foot state of the humanoid robot. The first layer uses the Generalized Type 2 Fuzzy C-means Algorithm (GT2FCM) to calculate the probability of foot contact with the outside world based on the triaxial acceleration and angular velocity, detecting contact events that ensure the foot is standing. The second layer uses the Interval Type 2 Fuzzy Inference System (IT2FIS) to detect lateral sliding events of the foot based on the contact probability, linear velocity, and linear acceleration.

2. The robot foot contact and sliding detection method based on fuzzy logic according to claim 1, characterized in that, The method employs a proprioceptive inertial measurement unit (IMU) installed on the humanoid robot's foot to collect the three-axis acceleration and three-axis angular velocity of the humanoid robot's foot. This includes: collecting data at a frequency of 1000 Hz, wherein the data contains Gaussian white noise, with the accelerometer Gaussian white noise being 0.01 m / s² to 0.05 m / s², and the gyroscope random walk angular frequency being 0.0001 radians per second per square root hertz to 0.001 radians per second per square root hertz.

3. The robot foot contact and sliding detection method based on fuzzy logic according to claim 2, characterized in that, The first layer employs the Generalized Type 2 Fuzzy C-means Algorithm (GT2FCM) to calculate the probability of foot contact with the outside world based on the foot's triaxial acceleration and triaxial angular velocity. This includes: classifying the input and output data using a subtractive clustering algorithm to determine the cluster centers and their number r; constructing a fuzzy rule matrix R and a second-level membership matrix SM, where the fuzzy rule matrix R contains r fuzzy rules, each corresponding to the uncertainty value of the d-dimensional input and output variables, and the second-level membership matrix SM represents the weights of each uncertainty dimension; calculating the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM; using the GT2FCM algorithm to calculate the membership degree between the input data and each fuzzy rule; and normalizing the membership degree to obtain the contact probability PO.

4. The robot foot contact and sliding detection method based on fuzzy logic according to claim 3, characterized in that, The calculation of the type restoration matrix UNR based on the fuzzy rule matrix R and the second-level membership matrix SM includes: ； ； ； ； in, Let be the weighted representative value of the j-th variable in fuzzy rule i under uncertainty condition g, obtained from the second-level membership matrix SM. This refers to the element in the i-th row and m-th column of the SM matrix; Let be the i-th row and j-th column of the m-th two-dimensional matrix in the R three-dimensional matrix; Input elements for uncertainty; Output the r-th rule for uncertainty type reduction; These are the input elements of the training set; Output elements for the training set.

5. The robot foot contact and sliding detection method based on fuzzy logic according to claim 4, characterized in that, The second layer employs an Interval Type 2 Fuzzy Inference System (IT2FIS) to detect lateral foot sliding events based on the contact probability between the foot and the external environment, the linear velocity of the foot, and the linear acceleration of the foot. This includes input data such as the contact probability PO, the x-axis foot velocity and acceleration, and the y-axis foot velocity and acceleration. The time window of the Interval Type 2 Fuzzy Inference System (IT2FIS) is matched to the robot's operating speed; the time window decreases as the speed increases. Fuzzy inference distinguishes between contact events that ensure foot standing and lateral foot sliding events.

6. The robot foot contact and sliding detection method based on fuzzy logic according to claim 5, characterized in that, The calculation of the probability of foot contact with the external environment based on the foot's triaxial acceleration and triaxial angular velocity also includes: ； ； ； in, The velocity of the foot along the z-axis; The acceleration at the foot along the z-axis; Let f be the rate of change of foot acceleration along the z-axis; f is the robot foot; based on the foot velocity and acceleration along the x-axis, foot velocity and acceleration along the y-axis, and the three-axis angular velocity of the foot, the probability of foot contact with the outside world is generated using the generalized type 2 fuzzy C-means algorithm GT2FCM.

7. The robot foot contact and sliding detection method based on fuzzy logic according to claim 6, characterized in that, After detecting the lateral sliding event of the foot, the method further includes: using the multi-joint dynamics and contact MuJoCo simulation environment to simulate and verify the open-source humanoid robot. The verification set includes walking backwards on flat ground, walking with a predetermined stride, walking on unstructured ground, and walking up stairs, covering both structured and unstructured terrain.

8. A robot foot contact and sliding detection device based on fuzzy logic, characterized in that, include: The first main module is used to collect the triaxial acceleration and angular velocity of the humanoid robot's feet using a proprioceptive inertial measurement unit (IMU) installed on the robot's feet. The second main module performs Kalman filtering on the triaxial acceleration and angular velocity of the humanoid robot's feet to obtain the linear acceleration, linear velocity, and angular velocity of the humanoid robot's feet in the world coordinate system. The third main module uses hierarchical fuzzy logic to detect the foot state of the humanoid robot. The first layer uses the Generalized Type 2 Fuzzy C-means Algorithm (GT2FCM) to calculate the probability of foot contact with the outside world based on the triaxial acceleration and angular velocity, and detects contact events that ensure the foot is standing. The fourth main module uses the Interval Type 2 Fuzzy Inference System (IT2FIS) in the second layer to detect lateral sliding events of the foot based on the probability of foot contact with the outside world, the linear velocity, and the linear acceleration.

9. An electronic device, characterized in that, include: At least one processor, at least one memory, and a communication interface; wherein, The processor, memory, and communication interface communicate with each other; The memory stores program instructions that can be executed by the processor, which invokes the program instructions to perform the method described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions that cause the computer to perform the method described in any one of claims 1 to 7.

Images