CN112947065A - AZR (AZR) adjusting method for walking real-time gait of biped robot - Google Patents

AZR (AZR) adjusting method for walking real-time gait of biped robot Download PDF

Info

Publication number
CN112947065A
CN112947065A CN202110100304.5A CN202110100304A CN112947065A CN 112947065 A CN112947065 A CN 112947065A CN 202110100304 A CN202110100304 A CN 202110100304A CN 112947065 A CN112947065 A CN 112947065A
Authority
CN
China
Prior art keywords
azr
robot
zmp
walking
gait
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.)
Granted
Application number
CN202110100304.5A
Other languages
Chinese (zh)
Other versions
CN112947065B (en
Inventor
卢志强
侯媛彬
孟芸
柴秀丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Henan University
Original Assignee
Henan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Henan University filed Critical Henan University
Priority to CN202110100304.5A priority Critical patent/CN112947065B/en
Publication of CN112947065A publication Critical patent/CN112947065A/en
Application granted granted Critical
Publication of CN112947065B publication Critical patent/CN112947065B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Abstract

The invention relates to an AZR (autoregressive moving distance) adjusting method for walking real-time gaits of a biped robot, which comprises the steps of constructing an online database by using measured and calculated data of an offline system of the robot, wherein the online database comprises a robot step size set S, AZR set H, a walking gaits set G and an energy consumption set E, and the distance d of a target travel of a given robot and an expected AZR position r are measured at the given robotAZRPlanning the step length sequence with the lowest energy consumption
Figure DDA0002913319290000011
Each step length of the robot is taken out
Figure DDA0002913319290000012
And AZR variable ηiInquiring an on-line database to obtain a motor angle sequence g for gait controliAccording to the pressure set F of the walking steps of the robotiCalculating the real-time ZMP trajectory rZMPBy using rZMP(n) and rAZRDeviation e of (n)iObtaining AZR variable eta by PI correction methodiCorrection value η ofi+1And according to the correction value etai+1And step length si+1And inquiring an online database, and optimizing the gait track of the biped robot online. The invention can overcome the interference of robot modeling and environmental errors, realize gait control with high robustness and low energy consumption in the walking process of the biped robot, and better solve the walking problem of the biped robot with highly nonlinear characteristics.

Description

AZR (AZR) adjusting method for walking real-time gait of biped robot
Technical Field
The invention relates to the field of motion design of biped robots, in particular to an AZR (autoregressive distance) adjusting method for walking real-time gait of a biped robot.
Background
Because the walking gait of the biped robot (humanoid robot) has a highly nonlinear characteristic, a modeling error inevitably exists in the modeling of the biped robot, and in order to meet the requirement of robustness, the prior method fixes the ZMP (Zero Moment Point, ZMP) track of the robot motion at the center of the supporting leg to realize the most stable walking, but the method is not an efficient method, so that the walking consumes more energy;
through research on an AZR (Allowable ZMP Region, AZR) method, the method is a better method for overcoming the balance between modeling errors and efficient walking of the robot, some edge regions are marked in the regions of the supporting legs to compensate the modeling errors, and the ZMP track of the walking of the robot is positioned in the AZR in the middle of the supporting legs;
the stability or energy consumption of the biped robot is related to a variable eta describing the AZR area size, if the AZR variable eta can be dynamically adjusted when the robot travels, the robot can obtain the optimal gait plan when the robot travels, and therefore a better compromise is obtained between the energy consumption and the stability.
Disclosure of Invention
The invention provides an AZR (amplitude-variation-ratio) adjusting method for walking real-time gait of a biped robot for realizing online optimization of walking of the biped robot
Figure BDA0002913319270000011
And passes through the AZR position r during the robot walkingAZR(n) actual gait position r of biped robotZMPDeviation e between (n)iUpdating by PI correctionVariable η of AZRiAnd further according to the variable η of AZRi+1And step length si+1And inquiring an online database to adjust the gait plan of the robot online. The interference of robot modeling and environmental errors is overcome, gait control with high robustness and low energy consumption is realized in the walking process of the biped robot, and the walking problem of the biped robot with highly nonlinear characteristics is well solved.
The invention provides an AZR (AZR) adjusting method for walking real-time gait of a biped robot, which comprises the following steps:
step l: the method comprises the steps that an online database is built by using measurement and calculation data of a robot offline system, wherein the online database comprises a robot step size set S, AZR set H, a walking gait set G and an energy consumption set E;
step 2: given a robot target travel distance d and a desired AZR position rAZRPlanning the step length sequence with the lowest energy consumption
Figure BDA0002913319270000012
And step 3: each step length of the robot is taken out
Figure BDA0002913319270000013
And AZR variable ηiInquiring an on-line database to obtain a motor angle sequence g for gait controli；
And 4, step 4: according to the pressure set F of the steps during the walking of the robotiCalculating the real-time ZMP trajectory rZMP(n) based on rZMP(n) and rAZR(n) obtaining a deviation e in the Y-axis directioni；
And 5: using the deviation e of the Y-axis directioniObtaining AZR variable eta by PI correction methodiCorrection value η ofi+1And according to said correction value etai+1And step length si+1And inquiring an online database, and optimizing the gait track of the biped robot online.
Further, step 2 specifically includes:
step 2.1: the step length sequence comprises a start step, an end step and a plurality of loop steps smA plurality of said cyclic steps smStep length is equal, and according to the motion rule of the robot, the initial step is defined to comprise s1And s2Said stopping step comprising sc-1And scLet s1＝sc、s2＝sc-1And the total length of the start step and the stop step is 1-2 times of the cycle step smStep size of (2);
step 2.2: given the robot target travel distance d, the cycle step s is knownmIs calculated by the formula (1) said start step s1And s2Step length:
Figure BDA0002913319270000021
wherein d isbThe total length of the start step and the stop step; c is the total number of steps;
step 2.3: given the expected AZR position rAZRTaking a cyclic step S from the set SmForming a step sequence SmSaid step length sequence SmThe energy consumption function of (2);
Figure BDA0002913319270000022
wherein eta is AZR position rAZRCorresponding to the value of the variable of the area AZR,
Figure BDA0002913319270000023
indicating the step of the cycle smThe energy consumption of (2) is reduced,
Figure BDA0002913319270000024
indicates the start of step s1The energy consumption of (2) is reduced,
Figure BDA0002913319270000025
indicating a stop step scThe energy consumption of (2) is reduced,
Figure BDA0002913319270000026
and
Figure BDA0002913319270000027
the equivalent cycle step length is s when the moving distance of the swing foot in front of the body is unequal to the moving distance of the swing foot in the process of starting and stopping the robot walking1And s2Energy consumption of (2);
step 2.4: combining the formula (2) to obtain the optimal cycle step length
Figure BDA0002913319270000028
The optimal cycle step size
Figure BDA0002913319270000029
Correspondingly forming the step length sequence with the lowest energy consumption
Figure BDA00029133192700000210
The step length sequence with the lowest energy consumption
Figure BDA00029133192700000211
The constraint conditions shown in formula (3) are satisfied;
Figure BDA00029133192700000212
wherein the content of the first and second substances,
Figure BDA00029133192700000213
denotes smThe energy consumption value calculated when the kth value in the step set S is taken, k is 1, 2, …, L, and L is the number of elements in the step set S.
Further, step 4 specifically includes:
step 4.1 the real-time ZMP trajectory rZMP(n) is represented by:
rZMP(n)＝[xZMP(n) yZMP(n) 0]T (4)；
wherein x and y respectively represent the front and side directions of the robot;
combining the formula (4), the robot walks the ith step, and the value set F is obtained by the robot sole pressure sensor at the sampling point n timei(n) calculating the real-time ZMP trajectory rZMP(n) represented by formula (5):
Figure BDA0002913319270000031
wherein the content of the first and second substances,
Figure BDA0002913319270000032
and
Figure BDA0002913319270000033
the position and pressure of the jth sensor in the x-axis direction and the y-axis direction, respectively, cnIs the number of sensors.
Further, the step 4 further includes:
step 4.2: get biped robot per step time TSFor controlling the period, the biped robot has a step time TSExpressed as:
TS＝N·ts (6)；
wherein N is the gait cycle of the biped robot;
step 4.3: defining the duty cycle of the robot motion as sigma ≡ 2N1The motion state of the robot is judged according to the value of the sampling point N;
if N is present1＜n≤N2The biped robot is in a single support stage SSP;
if N is more than or equal to 1 and less than or equal to N1，N2N is less than or equal to N, and the biped robot is in a DSP (digital signal processor) of a double-support stage;
wherein N is1＝σN/2，N2＝N-N1，n＝1，2，…，N；
Step 4.4: when the biped robot is in the DSP stage of dual support, rZMP(n) is located in the desired AZR region;
when the biped robot is in the single support stage SSP, the X-axis direction X of the movement is requiredZMP(n) monotonically increases, and xZMP(N1+1)≥xAZR(N1+1)、xZMP(N2)≤xAZR(N2)。
Further, combining with the formula (5), the deviation value e of the Y-axis direction in step 4iExpressed as:
Figure BDA0002913319270000034
wherein, yAZRWhen (n) is not less than 0, co＝1；yAZRWhen (n) < 0, co＝-1，lfwThe robot is wide enough.
Further, the step 5 specifically includes:
step 5.1: establishing a PI model with an incremental transfer function, wherein the AZR variable etaiCorrection value η ofi+1As expressed by equation (8):
ηi+1＝ηi+Δηi+1 (8)；
wherein, the increment is delta etai+1As expressed by equation (9):
Figure BDA0002913319270000041
wherein k isPIs the proportionality coefficient, TIIs the integration time constant, TSIs the control period;
step 5.2: according to said correction value etai+1And step length si+1Inquiring an online database to obtain a gait track g for controlling the biped robot to walk at the (i + 1) th stepi+1，gi+1As expressed by equation (10):
Figure BDA0002913319270000042
wherein q is the angle of the joint motor of the robot, m is the number of the joint motors, and N is the gait cycle.
Further, the step 5 further comprises:
at the calculation of Δ ηi+1When introducing a time constant of TLFirst order inertia element of
Figure BDA0002913319270000043
Smoothing is carried out:
Δηi+1＝KαΔηi+K1ei+K2ei-1 (11)；
wherein the content of the first and second substances,
Figure BDA0002913319270000044
through the technical scheme, the invention has the beneficial effects that:
the invention is provided with an online database which comprises a biped robot step length set S, AZR set H, a walking gait set G and an energy consumption set E, wherein an element S of the set S is combined with an element eta in the set H, the element G of the G in the walking gait set, namely G ← (S, eta), is inquired, a motor angle for controlling the robot to walk is stored in the element G, and the energy consumption is an element of the energy consumption set E
Figure BDA0002913319270000045
When the travel distance d and the AZR position r of the robot target are givenAZRAnd then planning the step length sequence with the lowest energy consumption of the robot
Figure BDA0002913319270000046
In the step length sequence
Figure BDA0002913319270000047
Step length s of the ith step of robot walking is takeniAnd corresponding AZR variable ηiE.g. H, inquiring an online database to obtain a gait track g of the angle of the robot joint motoriIn the walking process of the biped robot, due to factors such as modeling errors and environmental changes, gait can be deviated, and the real-time ZMP trajectory r is calculatedZMPObtaining and expecting AZR position rAZRDeviation value e ofiObtaining AZR variable eta by PI correction methodiCorrection value η ofi+1According to said correction value ηi+1And step length si+1Query online database, online optimize biped machineHuman gait trajectory.
The method for adjusting the AZR of the real-time gait of the biped robot can perform compromise calculation on the stability and the energy consumption of the biped robot in the walking control of the biped robot with highly nonlinear characteristics, and realize that the gait of the biped robot keeps stable walking under the condition of lowest energy consumption by performing online adjustment on an AZR variable eta.
Drawings
FIG. 1 is a flow chart of the AZR adjustment method for real-time walking gait of a biped robot of the present invention;
FIG. 2 is a system working principle diagram of the AZR adjusting method for real-time walking gait of the biped robot.
Fig. 3 is a schematic diagram of the relationship between the AZR region and the robot support region of the AZR adjustment method for real-time walking gait of the biped robot of the present invention.
Fig. 4 is a motion energy consumption analysis diagram of robot gait planning of the AZR adjusting method of walking real-time gait of the biped robot of the invention.
FIG. 5 is a dynamic simulation diagram of robot gait movement in the AZR adjustment method of walking real-time gait of the biped robot of the present invention.
Fig. 6 is a schematic diagram of deviation value calculation in the AZR adjustment method of walking real-time gait of the biped robot of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
As shown in fig. 1 and 2, an AZR adjusting method for a walking real-time gait of a biped robot according to an embodiment of the present invention includes the following steps:
step 1: the method comprises the steps that an online database is built by using measurement and calculation data of a robot offline system, wherein the online database comprises a robot step size set S, AZR set H, a walking gait set G and an energy consumption set E;
step 2: given a robot target travel distance d and a desired AZR position rAZRPlanning the step length sequence with the lowest energy consumption
Figure BDA0002913319270000051
And step 3: each step length of the robot is taken out
Figure BDA0002913319270000052
And AZR variable ηiInquiring an on-line database to obtain a motor angle sequence g for gait controli；
And 4, step 4: according to the pressure set F of the steps during the walking of the robotiCalculating the real-time ZMP trajectory rZMP(n) based on rZMP(n) and rAZR(n) obtaining a deviation e in the Y-axis directioni；
And 5: using the deviation e of the Y-axis directioniObtaining AZR variable eta by PI correction methodiCorrection value η ofi+1And according to said correction value etai+1And step length si+1And inquiring an online database, and optimizing the gait track of the biped robot online.
The method comprises the steps of constructing an online database, wherein the online database comprises a biped robot step size set S, AZR set H, a walking gait set G and an energy consumption set E, combining an element S of the set S with an element eta in the set H, and inquiring an element G of the G in the walking gait set, namely G ← (S, eta), wherein the element G stores a motor angle for controlling the robot to walk, and the energy consumption is an element of the energy consumption set E
Figure BDA0002913319270000061
Combining the expected AZR position r after the travel distance of the robot target is given by using the online databaseAZRPlanning the step length sequence with the lowest energy consumption
Figure BDA0002913319270000062
And make the robot follow the step length sequence
Figure BDA0002913319270000063
When the robot walks with deviation due to interference of factors such as environment and the like, the method calculates r in the walking process of the robot in real timeZMP(n) track and sum position rAZRDeviation e between (n)iObtaining variable eta of AZR by PI correction methodi+1Combined step length si+1And variable η of AZRi+1And inquiring the online database again, and regulating the gait of the biped robot in the advancing process to complete the online gait optimization of the biped robot.
Example 2
As shown in fig. 4 and 5, based on the above embodiment 1, in this embodiment, the step sequence with the lowest energy consumption is obtained
Figure BDA0002913319270000068
And optimizing the step 2, specifically:
step 2.1: the step length sequence comprises a start step, an end step and a plurality of loop steps smA plurality of said cyclic steps smStep length is equal, and according to the motion rule of the robot, the initial step is defined to comprise s1And s2Said stopping step comprising sc-1And scLet s1＝sc、s2＝sc-1And the total length of the start step and the stop step is 1-2 times of the cycle step smStep size of (2);
step 2.2: given the robot target travel distance d, the cycle step s is knownmIs calculated by the formula (1) said start step s1And s2Step length:
Figure BDA0002913319270000064
wherein d isbThe total length of the start step and the stop step; c is the total number of steps;
step 2.3: given the expected AZR position rAZRTaking a cyclic step S from the set SmForming a step sequence SmSaid step length sequence SmThe energy consumption function of (2);
Figure BDA0002913319270000065
wherein eta is AZR position rAZRCorresponding to the value of the variable of the area AZR,
Figure BDA0002913319270000066
indicating the step of the cycle smThe energy consumption of (2) is reduced,
Figure BDA0002913319270000067
indicates the start of step s1The energy consumption of (2) is reduced,
Figure BDA0002913319270000071
indicating a stop step scThe energy consumption of (2) is reduced,
Figure BDA0002913319270000072
and
Figure BDA0002913319270000073
the equivalent cycle step length is s when the moving distance of the swing foot in front of the body is unequal to the moving distance of the swing foot in the process of starting and stopping the robot walking1And s2Energy consumption of (2);
step 2.4: combining the formula (2) to obtain the optimal cycle step length
Figure BDA0002913319270000074
The optimal cycle step size
Figure BDA0002913319270000075
Correspondingly forming the step length sequence with the lowest energy consumption
Figure BDA0002913319270000076
The step length sequence with the lowest energy consumption
Figure BDA0002913319270000077
The constraint conditions shown in formula (3) are satisfied;
Figure BDA0002913319270000078
wherein the content of the first and second substances,
Figure BDA0002913319270000079
denotes smThe energy consumption value calculated when the kth value in the step set S is taken, k is 1, 2, …, L, and L is the number of elements in the step set S.
Example 3
Based on the above embodiment 1, in the present embodiment, to obtain the real-time ZMP trajectory rZMP(n) the step 4 is optimized, and specifically comprises the following steps:
step 4.1 the real-time ZMP trajectory rZMP(n) is represented by:
rZMP(n)＝[xZMP(n) yZMP(n) 0]T (4)；
wherein x and y respectively represent the front and side directions of the robot;
combining the formula (4), the robot walks the ith step, and the value set F is obtained by the robot sole pressure sensor at the sampling point n timei(n) calculating the real-time ZMP trajectory rZMP(n) represented by formula (5):
Figure BDA00029133192700000710
wherein the content of the first and second substances,
Figure BDA00029133192700000711
and fi j(n)∈Fi(n) the position and pressure of the jth sensor in the x-axis direction and the y-axis direction, respectively, cnIs the number of sensors.
Example 4
As shown in fig. 6, based on the above-mentioned embodiment 3, in this embodimentIn the embodiment, the deviation e in the Y-axis direction is calculatediAnd optimizing the step 4, specifically:
step 4.2: get biped robot per step time TSFor controlling the period, the biped robot has a step time TSExpressed as:
TS＝N·ts (6)；
wherein N is the gait cycle of the biped robot;
step 4.3: defining the duty cycle of the robot motion as sigma ≡ 2N1The motion state of the robot is judged according to the value of the sampling point N;
if N is present1＜n≤N2The biped robot is in a single support stage SSP;
if N is more than or equal to 1 and less than or equal to N1，N2N is less than or equal to N, and the biped robot is in a DSP (digital signal processor) of a double-support stage;
wherein N is1＝σN/2，N2＝N-N1，n＝1，2，…，N；
As an alternative embodiment, the foot length l is set as shown in FIG. 3flWide foot lfwThe biped robot has a stride of s and a biped Y-axis distance of w. For the example of the Left Foot (LF) in front and the Right Foot (RF) in back, in the first dual support stage DSP, AZR is defined by point r1、r2、r3、r4、r5、r6Hexagonal, then in single support SSP, AZR is the number of points r4、r5、r6、r7A rectangular area is formed, in the second dual-support stage DSP, AZR is formed by point r5、r6、r7、r8、r9、r10A hexagon is formed;
are used separately
Figure BDA0002913319270000081
And
Figure BDA0002913319270000082
representing the percentage of AZR in the foot length and foot width directions, gait planning in this exampleIn the algorithm, the algorithm is carried out,
Figure BDA0002913319270000083
taking a fixed value and appointing a variable describing the AZR area
Figure BDA0002913319270000084
Step 4.4: when the biped robot is in the DSP stage of dual support, rZMP(n) is located in the desired AZR region;
when the biped robot is in the single support stage SSP, the X-axis direction X of the movement is requiredZMP(n) monotonically increases, and xZMP(N1+1)≥xAZR(N1+1)、xZMP(N2)≤xAZR(N2)。
As an alternative embodiment, as shown in FIG. 6, the deviation e of the Y-axis direction in step 4 is combined with the formula (5)iExpressed as:
Figure BDA0002913319270000085
wherein, yAZRWhen (n) is not less than 0, co＝1；yAZRWhen (n) < 0, co＝-1，lfwThe robot is wide enough.
Example 5
In addition to the above embodiments, the embodiment of the present invention is different from the above embodiments in that the deviation e in the Y-axis direction is obtainediThen, the deviation e in the Y-axis direction is calculatediAs feedback value, η to the variable AZRiCorrecting, and inquiring the online database again to complete gait updating, which specifically comprises the following steps:
step 5.1: establishing a PI model with an incremental transfer function, wherein the AZR variable etaiCorrection value η ofi+1As expressed by equation (8):
ηi+1＝ηi+Δηi+1 (8)；
wherein, the increment is delta etai+1As expressed by equation (9):
Figure BDA0002913319270000091
wherein k isPIs the proportionality coefficient, TIIs the integration time constant, TSIs the control period;
step 5.2: according to said correction value etai+1And step length si+1Inquiring an online database to obtain a gait track g for controlling the biped robot to walk at the (i + 1) th stepi+1，gi+1As expressed by equation (10):
Figure BDA0002913319270000092
wherein q is the angle of the joint motor of the robot, m is the number of the joint motors, and N is the gait cycle.
As an implementation manner, the step 5 further includes:
at the calculation of Δ ηi+1When introducing a time constant of TLFirst order inertia element of
Figure BDA0002913319270000093
Smoothing is carried out:
Δηi+1＝KαΔηi+K1ei+K2ei-1 (11)；
wherein the content of the first and second substances,
Figure BDA0002913319270000094
the method comprises the steps of constructing an online database, wherein the online database comprises a biped robot step size set S, AZR set H, a walking gait set G and an energy consumption set E, combining an element S of the set S with an element eta in the set H, and inquiring an element G of the G in the walking gait set, namely G ← (S, eta), wherein the element G stores a motor angle for controlling the robot to walk, and the energy consumption is an element of the energy consumption set E
Figure BDA0002913319270000095
Combining the expected AZR position r after the travel distance of the robot target is given by using the online databaseAZRPlanning the step length sequence with the lowest energy consumption
Figure BDA0002913319270000096
And make the robot follow the step length sequence
Figure BDA0002913319270000097
When the robot walks with deviation due to interference of factors such as environment and the like, the method calculates r in the walking process of the robot in real timeZMP(n) track and sum position rAZRDeviation e between (n)iObtaining variable eta of AZR by PI correction methodi+1Combined step length si+1And variable η of AZRi+1And thirdly, inquiring an online database, and regulating and adjusting the gait of the biped robot during the running process to complete the online gait optimization of the biped robot.
The following experiments were conducted to demonstrate the effects of the present invention
In experimental studies on biped robots, it is necessary to suggest that biped robot gait walking is based on the following assumptions:
(1) the biped robot body maintains an upright posture at all times. When a person walks, the pitching angle of the trunk is generally within 3 degrees, and most researches generally accept the assumption that the body of the biped robot is upright in the walking process.
(2) The feet of the biped robot are always parallel to the ground. Most of the common humanoid robots do not have toes, and cannot play a role in improving the driving performance when lifting and lowering feet.
(3) One gait cycle T of the biped robot, including a Double Support Phase (DSP) time TDSPAnd Single Support (SSP) time TSSPDefining the duty cycle of the DSP
Figure BDA0002913319270000101
When a person walks, the σ interval is about 15% to 25%, and σ is 25% in the algorithm of the present embodiment.
In the experiment, the target travel distance d is 100cm, and is taken
Figure BDA0002913319270000102
The step length sequence S is carried out according to the formulas (1) and (2)*The result of the planning is shown in fig. 4, where η is 0.2, 0.3, …, 1, when the loop step s is repeatedmWhen the distance is 10cm, the energy consumption consumed by the robot movement is the minimum in the same row of data;
at this time, the total number of steps c is calculated to be 12, and the step s is started1＝3cm、s27cm, stop step sc-1、scCorresponding to the step length sequence S with the lowest energy consumption of 7cm and 3cmm＝{3，7，10，10，10，10，10，10，10，10，7，3}。
By calculation, when the variable eta of AZRiSetting the robot to be a double-support stage DSP with two feet standing together, wherein the swing length L of the right foot is 0.8rfLeft foot swing length L, {3, 17, 20, 20, 17, 3}, and left foot swing length Llf＝{10，20，20，20，20，10}；
Swing length L of right footrfThe middle two 3cm respectively correspond to the start step and the stop step, and the energy consumption is respectively
Figure BDA0002913319270000103
Figure BDA0002913319270000104
Swing length L of right footrfMiddle 17cm and left foot swing length LlfThe middle 10cm represents the condition that the movement distance of the swinging foot is unequal at the front and the back of the supporting foot,
the total energy consumption for the robot to travel is obtained by the formula (2)
Figure BDA0002913319270000105
Wherein
Figure BDA0002913319270000106
Correspondingly taking each step g of the robot from the databasei＝[q1 q2 … q16]Controlling the joint motion of the robot, wherein i is 1, 2…, 12 step 13 is added to make the robot stop moving with feet in parallel, the dynamic simulation of the robot gait motion is shown in figure 5, the black point is the body mass center rb(n) a motion trajectory.
According to the simulation data, performing actual measurement walking experiment on the robot;
in the experiment, the target stroke distance d is set to 100cm, and the AZR initial variable is set
Figure BDA0002913319270000108
TS＝1.6s，kP＝0.875，TI＝3.2s，TL1.6s, in the single-support phase SSP, along the boundary line 0.8 of the AZR variable η;
the initial condition of the robot walking is e0＝e-1＝0，
Figure BDA0002913319270000107
The eta value of the AZR controller is stabilized at 0.8 after the third step of robot walking, and the correction calculation data in the period are shown in the table 1:
TABLE 1 variable value and energy consumption of biped robot walking 100cm
Figure BDA0002913319270000111
In table siThe data in the parentheses correspond to the movement distances of the swing legs in the back and front of the body respectively,
Figure BDA0002913319270000112
to correspond to the actual ηiThe energy consumption of the calculation is calculated,
Figure BDA0002913319270000113
is composed of
Figure BDA0002913319270000114
Energy consumption value of (2). Calculate to obtain the reality
Figure BDA0002913319270000115
While
Figure BDA0002913319270000116
I.e. AZR controller regulation eta during walkingiThe energy consumption is reduced by 12.13 percent.
The above-described embodiments are merely preferred embodiments of the present invention, and not intended to limit the scope of the invention, so that equivalent changes or modifications in the structure, features and principles described in the present invention should be included in the claims of the present invention.

Claims (7)

1. An AZR adjusting method for walking real-time gait of a biped robot is characterized by comprising the following steps:
step 1: the method comprises the steps that an online database is built by using measurement and calculation data of a robot offline system, wherein the online database comprises a robot step size set S, AZR set H, a walking gait set G and an energy consumption set E;
step 2: given a robot target travel distance d and a desired AZR position rAZRPlanning the step length sequence with the lowest energy consumption
Figure FDA0002913319260000014
And step 3: each step length of the robot is taken out
Figure FDA0002913319260000013
And AZR variable ηiInquiring an on-line database to obtain a motor angle sequence g for gait controli；
And 4, step 4: according to the pressure set F of the steps during the walking of the robotiCalculating the real-time ZMP trajectory rZMP(n) based on rZMP(n) and rAZR(n) obtaining a deviation e in the Y-axis directioni；
And 5: using the deviation e of the Y-axis directioniObtaining AZR variable eta by PI correction methodiCorrection value η ofi+1And according to said correction value etai+1And step length si+1And inquiring an online database, and optimizing the gait track of the biped robot online.
2. The AZR adjusting method for the walking real-time gait of the biped robot according to claim 1, characterized in that the step 2 specifically comprises:
step 2.1: the step length sequence comprises a start step, an end step and a plurality of loop steps smA plurality of said cyclic steps smStep length is equal, and according to the motion rule of the robot, the initial step is defined to comprise s1And s2Said stopping step comprising sc-1And scLet s1＝sc、s2＝sc-1And the total length of the start step and the stop step is 1-2 times of the cycle step smStep size of (2);
step 2.2: given the robot target travel distance d, the cycle step s is knownmIs calculated by the formula (1) said start step s1And s2Step length:
Figure FDA0002913319260000011
wherein d isbThe total length of the start step and the stop step; c is the total number of steps;
step 2.3: given the expected AZR position rAZRTaking a cyclic step S from the set SmForming a step sequence SmSaid step length sequence SmThe energy consumption function of (2);
Figure FDA0002913319260000012
wherein eta is AZR position rAZRCorresponding to the value of the variable of the area AZR,
Figure FDA0002913319260000021
indicating the step of the cycle smThe energy consumption of (2) is reduced,
Figure FDA0002913319260000022
indicates the start of step s1The energy consumption of (2) is reduced,
Figure FDA0002913319260000023
indicating a stop step scThe energy consumption of (2) is reduced,
Figure FDA0002913319260000024
and
Figure FDA0002913319260000025
the equivalent cycle step length is s when the moving distance of the swing foot in front of the body is unequal to the moving distance of the swing foot in the process of starting and stopping the robot walking1And s2Energy consumption of (2);
step 2.4: combining the formula (2) to obtain the optimal cycle step length
Figure FDA0002913319260000026
The optimal cycle step size
Figure FDA0002913319260000027
Correspondingly forming the step length sequence with the lowest energy consumption
Figure FDA0002913319260000028
The step length sequence with the lowest energy consumption
Figure FDA0002913319260000029
The constraint conditions shown in formula (3) are satisfied;
Figure FDA00029133192600000210
wherein the content of the first and second substances,
Figure FDA00029133192600000211
friend show smThe energy consumption value calculated when the kth value in the step set S is taken, k is 1, 2, …, L, and L is the number of elements in the step set S.
3. The AZR adjusting method for the walking real-time gait of the biped robot according to claim 1, characterized in that the step 4 specifically comprises:
step 4.1 the real-time ZMP trajectory rZMP(n) is represented by:
rZMP(n)＝[xZMP(n) yZMP(n) 0]T (4)；
wherein x and y respectively represent the front and side directions of the robot;
combining the formula (4), the robot walks the ith step, and the value set F is obtained by the robot sole pressure sensor at the sampling point n timei(n) calculating the real-time ZMP trajectory rZMP(n) represented by formula (5):
Figure FDA00029133192600000212
wherein the content of the first and second substances,
Figure FDA00029133192600000213
and
Figure FDA00029133192600000214
the position and pressure of the jth sensor in the x-axis direction and the y-axis direction, respectively, cnIs the number of sensors.
4. The AZR adjustment method for walking real-time gait of a biped robot according to claim 3, characterized in that the step 4 further comprises:
step 4.2: get biped robot per step time TSFor controlling the period, the biped robot has a step time TSExpressed as:
TS＝N·ts (6)；
wherein N is the gait cycle of the biped robot;
step 4.3: defining the duty cycle of the robot motion as sigma ≡ 2N1N, based on the value of the sampling point NJudging the motion state of the robot;
if N is present1＜n≤N2The biped robot is in a single support stage SSP;
if N is more than or equal to 1 and less than or equal to N1，N2N is less than or equal to N, and the biped robot is in a DSP (digital signal processor) of a double-support stage;
wherein N is1＝σN/2，N2＝N-N1，n＝1，2，…，N；
Step 4.4: when the biped robot is in the DSP stage of dual support, rZMP(n) is located in the desired AZR region;
when the biped robot is in the single support stage SSP, the X-axis direction X of the movement is requiredZMP(n) monotonically increases, and xZMP(N1+1)≥xAZR(N1+1)、xZMP(N2)≤xAZR(N2)。
5. The AZR adjusting method for walking real-time gait of biped robot according to claim 4, characterized in that the deviation value e of the Y-axis direction in step 4 is combined with formula (5)iExpressed as:
Figure FDA0002913319260000031
wherein, yAZRWhen (n) is not less than 0, co＝1；yAZRWhen (n) < 0, co＝-1，lfwThe robot is wide enough.
6. The AZR adjustment method for walking real-time gait of a biped robot according to claim 1, characterized in that the step 5 specifically comprises:
step 5.1: establishing a PI model with an incremental transfer function, wherein the AZR variable etaiCorrection value η ofi+1As expressed by equation (8):
ηi+1＝ηi+Δηi+1 (8)；
wherein, the increment is delta etai+1As expressed by equation (9):
Figure FDA0002913319260000032
wherein k isPIs the proportionality coefficient, TIIs the integration time constant, TSIs the control period;
step 5.2: according to said correction value etai+1And step length si+1Inquiring an online database to obtain a gait track g for controlling the biped robot to walk at the (i + 1) th stepi+1，gi+1As expressed by equation (10):
Figure FDA0002913319260000041
wherein q is the angle of the joint motor of the robot, m is the number of the joint motors, and N is the gait cycle.
7. The AZR adjustment method for walking real-time gait of a biped robot according to claim 6, characterized in that said step 5 further comprises:
at the calculation of Δ ηi+1When introducing a time constant of TLFirst order inertia element of
Figure FDA0002913319260000042
Smoothing is carried out:
Δηi+1＝KαΔηi+K1ei+K2ei-1 (11)；
wherein the content of the first and second substances,
Figure FDA0002913319260000043
CN202110100304.5A 2021-01-25 2021-01-25 AZR adjusting method for walking real-time gait of biped robot Active CN112947065B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110100304.5A CN112947065B (en) 2021-01-25 2021-01-25 AZR adjusting method for walking real-time gait of biped robot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110100304.5A CN112947065B (en) 2021-01-25 2021-01-25 AZR adjusting method for walking real-time gait of biped robot

Publications (2)

Publication Number Publication Date
CN112947065A true CN112947065A (en) 2021-06-11
CN112947065B CN112947065B (en) 2022-09-16

Family

ID=76236662

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110100304.5A Active CN112947065B (en) 2021-01-25 2021-01-25 AZR adjusting method for walking real-time gait of biped robot

Country Status (1)

Country Link
CN (1) CN112947065B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101847009A (en) * 2010-05-28 2010-09-29 广东工业大学 Biped robot gait energy efficiency optimization method
US20110153080A1 (en) * 2009-12-22 2011-06-23 Siemens Product Lifecycle Management Software Inc. Method and apparatus for industrial robotic pathscycle time optimization using fly by
CN103149933A (en) * 2013-02-27 2013-06-12 南京邮电大学 Closed-loop control-based humanoid robot omnidirectional walking method
CN108345211A (en) * 2017-01-23 2018-07-31 深圳市祈飞科技有限公司 Biped anthropomorphic robot and its non-linear gait planning method and control method
CN109164705A (en) * 2018-08-15 2019-01-08 重庆大学 A kind of dynamic bipod walking robot robust control method
CN109202901A (en) * 2018-08-29 2019-01-15 厦门理工学院 A kind of biped robot's stair climbing gait planning method, apparatus and robot
CN110315543A (en) * 2019-07-29 2019-10-11 北京理工大学 A kind of biped robot's gait generates and optimization method
CN110371213A (en) * 2019-07-12 2019-10-25 沈阳城市学院 A kind of biped robot's walking planning and control method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110153080A1 (en) * 2009-12-22 2011-06-23 Siemens Product Lifecycle Management Software Inc. Method and apparatus for industrial robotic pathscycle time optimization using fly by
CN101847009A (en) * 2010-05-28 2010-09-29 广东工业大学 Biped robot gait energy efficiency optimization method
CN103149933A (en) * 2013-02-27 2013-06-12 南京邮电大学 Closed-loop control-based humanoid robot omnidirectional walking method
CN108345211A (en) * 2017-01-23 2018-07-31 深圳市祈飞科技有限公司 Biped anthropomorphic robot and its non-linear gait planning method and control method
CN109164705A (en) * 2018-08-15 2019-01-08 重庆大学 A kind of dynamic bipod walking robot robust control method
CN109202901A (en) * 2018-08-29 2019-01-15 厦门理工学院 A kind of biped robot's stair climbing gait planning method, apparatus and robot
CN110371213A (en) * 2019-07-12 2019-10-25 沈阳城市学院 A kind of biped robot's walking planning and control method
CN110315543A (en) * 2019-07-29 2019-10-11 北京理工大学 A kind of biped robot's gait generates and optimization method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
周江琛: "基于变质心高度策略的双足机器人变步长步态规划", 《中南大学学报(自然科学版)》 *

Also Published As

Publication number Publication date
CN112947065B (en) 2022-09-16

Similar Documents

Publication Publication Date Title
CN110315543B (en) Biped robot gait generation and optimization method
CN112051741B (en) Dynamic motion generation and control method for biped robot
Buss et al. Preliminary walking experiments with underactuated 3D bipedal robot MARLO
Li et al. Walking motion generation, synthesis, and control for biped robot by using PGRL, LPI, and fuzzy logic
CN101847009B (en) Biped robot gait energy efficiency optimization method
Graichen et al. Control design for a bionic kangaroo
CN105867377A (en) Automatic navigation control method of agricultural machine
Chew et al. Dynamic bipedal walking assisted by learning
Liu et al. Bipedal walking with dynamic balance that involves three-dimensional upper body motion
CN114248855B (en) Biped robot space domain gait planning and control method
CN108931988B (en) Gait planning method of quadruped robot based on central pattern generator, central pattern generator and robot
CN104898672B (en) A kind of optimal control method of Humanoid Robot Based on Walking track
CN107831667B (en) Tracking control method for compensating human-computer interaction force of rehabilitation walking training robot
Horvat et al. Model predictive control based framework for com control of a quadruped robot
CN110737195A (en) Biped robot walking foot placement point planning method and device based on speed control
Uyanık et al. Adaptive control of a spring-mass hopper
CN107703762B (en) Human-computer interaction force identification and control method of rehabilitation walking training robot
CN112947065B (en) AZR adjusting method for walking real-time gait of biped robot
Dong et al. On-line gait adjustment for humanoid robot robust walking based on divergence component of motion
JP5440152B2 (en) Legged robot and gait data generation method for it
CN110990769B (en) Attitude migration algorithm system suitable for multi-degree-of-freedom robot
Wu et al. Kid-size robot humanoid walking with heel-contact and toe-off motion
Machbub et al. Slope Balancing Strategy for Bipedal Robot Walking Based on Inclination Estimation Using Sensors Fusion.
Kim et al. A model predictive capture point control framework for robust humanoid balancing via ankle, hip, and stepping strategies
CN111736598A (en) Harvester path tracking control method and system based on adaptive neural network

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