CN114509936A - Exercise planning method, device and medium for online learning of exercise capacity - Google Patents

Exercise planning method, device and medium for online learning of exercise capacity Download PDF

Info

Publication number
CN114509936A
CN114509936A CN202210412955.2A CN202210412955A CN114509936A CN 114509936 A CN114509936 A CN 114509936A CN 202210412955 A CN202210412955 A CN 202210412955A CN 114509936 A CN114509936 A CN 114509936A
Authority
CN
China
Prior art keywords
motion
curvature
function
motion platform
track
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
CN202210412955.2A
Other languages
Chinese (zh)
Other versions
CN114509936B (en
Inventor
华炜
胡艳明
韩正勇
沈峥
冯高超
郭磊
高海明
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.)
Zhejiang Lab
Original Assignee
Zhejiang Lab
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 Zhejiang Lab filed Critical Zhejiang Lab
Priority to CN202210412955.2A priority Critical patent/CN114509936B/en
Publication of CN114509936A publication Critical patent/CN114509936A/en
Application granted granted Critical
Publication of CN114509936B publication Critical patent/CN114509936B/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
    • G05B13/0205—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system
    • G05B13/024—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention belongs to the field of motion control, and relates to a motion planning method, a device and a medium for online learning of motion ability, wherein the method comprises the following steps: step S1: initializing a curvature-rate function according to the kinematic and dynamic constraints of the motion platform; step S2: evaluating the track tracking capability of the motion platform in real time according to the actual running track and the expected track of the motion platform; step S3: updating a curvature-rate function in real time according to the tracking capability of the motion platform; step S4: the motion planner of the motion platform obtains a motion planning result of the motion platform according to the updated curvature-rate function; step S5: and the controller of the motion platform obtains a motion command according to the motion planning result and controls the motion platform to move according to the motion command. The method can lead the motion platform to adaptively improve the motion efficiency under the condition of good motion capability; meanwhile, under the condition of poor motion capability, the tracking accuracy is improved by self-adaptively reducing the speed.

Description

Exercise planning method, device and medium for online learning of exercise capacity
Technical Field
The invention belongs to the field of motion control, and relates to a motion planning method, a device and a medium for online learning of motor ability.
Background
The motion platforms such as robots have high requirements on motion efficiency, precision and stability of track tracking, and the motion platforms can exert their own motion capability to the maximum extent, and meanwhile, the problems of efficiency reduction or track tracking precision and stability reduction caused by excessively conservative or aggressive motion planning results are avoided. The motion capability of the motion platform is greatly influenced by the factors such as material, installation precision, motion model calibration precision, motion environment, power performance of the motion platform and the like. Thus, the motion capabilities of different motion platforms may be very different, and even the motion capabilities of the same motion platform at different times and environments may be very different. In order to take the motion efficiency and the trajectory tracking accuracy into consideration, the motion planning of the motion platform needs to consider the motion capability of the motion platform.
Chinese patent application publication No. CN108549328A, entitled "adaptive speed planning method and system", discloses a method and system for creating a speed constraint curve according to the constraints of a motion platform (including one or more of the following: maximum rotation speed of each axis, centripetal acceleration-curvature constraint rate, rate limited by preventing vibration and reducing impact, and user demand constraint rate), and generating a speed planning curve considering the capability of the motion platform. The patent requires that the limiting conditions are preset manually, and the set limiting conditions are kept unchanged during the movement of the motion platform. The mode can exert the actual motion capability of the motion platform to the maximum extent in a fixed environment when the motion performance of the motion platform is stable. When the environmental changes are large or the motion performance of the motion platform is greatly influenced by the motion time and power (e.g., electric quantity), the method has a large limitation.
Chinese patent application CN112873208A, entitled "method and apparatus for planning real-time motion of anti-noise and dynamically constrained robot", discloses a method for enabling a motion platform to consider tracking errors online during operation, so that the motion platform can still track an expected track with high precision under the condition of noise in a control system. The patent considers the acceleration equality constraints of the trajectory tracking error for robot design of a particular mathematical model. The track tracking error information is a PID controller structure taking the track tracking error as an input, and the information is used as a penalty term of acceleration, namely: the larger the trajectory tracking error, the smaller the acceleration is constrained to be. The method only utilizes historical track tracking error information to adjust the acceleration constraint condition of the current one-step motion planning on line, and cannot be used for a more common multi-step motion planning task which can obtain the optimal motion in a time dimension. Because, at future motion planning time points, no trajectory tracking error information has yet been generated that can be used to calculate the motion constraints.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides a motion planning method, a motion planning device and a motion planning medium, wherein the motion planning method, the motion planning device and the motion planning medium can be used for learning a curvature-rate function on line according to the motion capability evaluated in real time in the motion process of a motion platform, and adjusting the motion planning constraint condition of the motion platform in real time, so that a motion planning result giving consideration to both motion efficiency and trajectory tracking precision is obtained, and the specific technical scheme is as follows:
an exercise planning method for online learning of exercise capacity comprises the following steps:
step S1: initializing a curvature-rate function according to the kinematic and dynamic constraints of the motion platform, wherein the curvature-rate function is as follows: a rate constraint function with curvature as an argument;
step S2: evaluating the track tracking capability of the motion platform in real time according to the actual running track and the expected track of the motion platform;
step S3: updating a curvature-rate function in real time according to the tracking capability of the motion platform;
step S4: the motion planner of the motion platform obtains a motion planning result of the motion platform according to the updated curvature-rate function;
step S5: and the controller of the motion platform obtains a motion command according to the motion planning result and controls the motion platform to move according to the motion command.
Further, the step S1 is specifically:
initializing a curvature-rate function as V according to kinematic and dynamic constraints of the motion platform0(k) = sqrt (a/| K |), a being a coefficient relating to the kinetic constraint of the motion platform, K ∈ a, a representing the curvature constraint of the motion platform, a = [ K ] ], [ K ∈ amin, Kmax]，KminAnd KmaxRespectively, the minimum and maximum curvatures of the moving platform.
Further, the step S2 is specifically:
according to the actual running track r = { q) of the moving platform1，q2，…，qN}， q1~qNFor the actual running track point, N is the recorded historical track step size, and its expected track rd = { p =1，p2，…，pM}，p1~pMFinding the actual running track point q of the motion platform at the moment of the distance t from the expected track rd for the expected running track point and M for the step length of the expected tracktClosest locus point pnearestAnd according to qtAnd pnearestCalculating the tracking error et(ii) a Then according to the tracking error etEvaluating a trajectory tracking capability u of a motion platformtWherein u ist<0，utThe larger the track tracing capacity of the motion platform, the worse the track tracing capacity is.
Further, the trajectory tracking capability utThe expression of (a) is:
ut=-(kPet+kI∑t i=0ei+kD(et-et-1) kP, kI and kD are weights.
Further, the tracking error etThe specific expression of (A) is as follows: e.g. of the typet=cp
Figure DEST_PATH_IMAGE002
eposition+ch
Figure 530499DEST_PATH_IMAGE002
eheadingWherein c ispAnd chIs the weight; e.g. of the typepositionFor tracking position errors by said pnearestAnd q is as describedtCalculating the Euclidean distance; e.g. of the typeheadingTo track angular error, through the expression | hnearest-htCalculated, | hnearestIs a track point pnearestCourse angle of (h)tFor the track point p of the motion platformtThe course angle of (c).
Further, the step S3 is specifically:
if ut<ua1Then V ist(k)=Vt-1(k)+b1|ut|G1(||k-kt||)；
If ut>ua2Then V ist(k)=Vt-1(k)+b2|ut|G2(||k-kt||)；
If ua1≤ut≤ua2Then V ist(k)=Vt-1(k)；
Wherein u isa1To set the lower bound, u, of the range of trackability allowed for a moving platforma2To set an upper bound on the range of allowable trackability, b1，b2For learning rate, Vt(k) The curvature-rate function, V, obtained for the time t updatet-1(k) As a function of curvature-rate at time t-1, G1And G2Is | | | k-ktI is a function of an argument, A is a domain of the curvature-rate function, ktThe curvature of the track point closest to the motion platform at time t in the expected track is shown.
Further, the function G1Satisfies G1<0, updating the obtained Vt(k) Satisfy at k<0 is a non-increasing function, at k>0 is a non-decreasing function, and Vt(k)<V0(k) For any k e A,
G1（k，kt）=-exp{-||k-kt||2/(2sigma12) Exp is an exponential function, and sigma1 is a parameter;
the function G2Satisfies G2>0, making said updated Vt(k) Satisfy at k<0 is a non-increasing function, at k>0 is a non-decreasing function, and Vt(k)<V0(k) For any k ∈ A, a preferred G2（k，kt）= exp{-||k-kt||2/(2sigma22) Exp is an exponential function and sigma2 is a parameter.
Further, the step S4 is specifically:
according to the curvature-speed function V obtained by updating at the moment tt(k) Constructing a curvature-rate constraint term v ≦ V (k), wherein v is the expected rate at any time, and V (k) represents a curvature-rate function at any time;
and obtaining an expected trajectory sequence and an expected speed sequence which meet the curvature-speed constraint term, namely an action planning result, in the action planner according to the curvature-speed constraint term.
An exercise planning device capable of learning exercise capacity online comprises one or more processors and is used for realizing the exercise planning method capable of learning exercise capacity online.
A computer-readable storage medium, having stored thereon a program which, when executed by a processor, implements the exercise planning method for on-line learning of exercise capacity.
Compared with the prior art, the method does not need to set limit conditions manually in advance, only utilizes the kinematics and the dynamic constraint of the motion platform to initialize the curvature-rate function, and updates the curvature-rate function on line according to the real-time estimated motion capability (represented by the track tracking capability), so that the motion platform obtains a multi-step motion planning result which gives consideration to the motion efficiency and the track tracking precision, the motion platform can adaptively improve the motion efficiency under the condition of good motion capability, and the method is not limited to the motion platform of a specific model and is suitable for all motion platforms which need to efficiently and accurately finish motion.
Drawings
FIG. 1 is a flow chart of a method for exercise planning for online learning of exercise capacity in an embodiment of the present invention;
FIG. 2 is a data flow diagram of an exercise planning method for online learning of exercise capacity in an embodiment of the present invention;
FIG. 3a is a graph according to a function G in an embodiment of the present invention1Satisfies G1<The motion capability at 0 updates the schematic diagram of the curvature-rate function on line;
FIG. 3b is a graph according to function G in an embodiment of the present invention2Satisfies G2>The motion capability at 0 updates a schematic diagram of a curvature-rate function on line;
fig. 4 is a schematic structural diagram of an exercise planning apparatus for online learning of exercise capacity according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and technical effects of the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples.
The present embodiment describes the present invention with an unmanned vehicle as a specific example of the motion platform in the present invention, and it should be understood that the embodiments provided herein are only for explaining the invention and are not intended to limit the invention.
As shown in fig. 1 and fig. 2, in the motion planning method for online learning of motion capability of the present invention, first, a curvature-rate function is initialized according to the kinematic and dynamic constraints of a motion platform; then, estimating the track tracking capability of the motion platform on line according to the actual track and the expected track of the motion platform, and updating a curvature-rate function on line according to the track tracking capability of the motion platform; and finally, the motion planner of the motion platform takes the updated curvature-rate function as a constraint condition to obtain a motion planning result considering the track tracking capability.
More specifically, the method comprises the following steps:
step S1: initializing a curvature-rate function according to the kinematic and dynamic constraints of the motion platform;
according to the kinematics and dynamics constraint of the unmanned vehicle, initializing a rate constraint function with curvature as an independent variable, namely initializing a curvature-rate function;
the curvature-rate function comprises the optimal rate of the unmanned automobile when the unmanned automobile tracks the motion path points with different curvatures, and the curvature-rate function is a non-decreasing function when the curvature is less than zero and is a non-increasing function when the curvature is more than or equal to zero;
specifically, according to the curvature constraint of the unmanned automobile, determining the definition domain A = [ K ] of the curvature-speed functionmin, Kmax]Wherein A represents a curvature constraint, KminAnd KmaxRespectively representing the minimum and maximum curvatures of the moving platform;
initializing the mapping relation of the curvature-speed function to V according to the dynamic constraint of the unmanned vehicle0，V0(k) = sqrt (a/| k |), a is a coefficient related to the dynamic constraint with the unmanned vehicle as the motion platform of the embodiment, and a preferable V0(k) And (4) the formula of the motion platform unmanned automobile is = sqrt (9.8f/| k |), wherein for any k ∈ A, 9.8 is an approximate gravity acceleration value, and f is an adhesion coefficient of the motion platform unmanned automobile and a road surface.
Step S2: evaluating the track tracking capability of the motion platform in real time according to the actual running track and the expected track of the motion platform;
estimating the track tracking capability of the unmanned vehicle in real time according to the actual running track and the expected track, namely the reference track, of the unmanned vehicle;
the actual running track r of the unmanned automobile consists of a series of historical actual running track points { q }1，q2，…，qNDenotes, wherein, the historical travel track point qiIncluding information on the spatial position, course angle, curvature, etc. of the unmanned vehicle, qi∈{q1，q2，…，qNN is the recorded historical track step length, one preferred N =1, which means that only the track point q of the current driverless car is recordedt；
The desired trajectory rd is composed of a sequence of points p1，p2，…，pMDenotes thatIn, piIncluding information on the spatial position, course angle, curvature, etc. of the unmanned vehicle, pi∈{p1，p2，…，pMM is the step length of the expected track;
from a desired sequence of trajectory points { p1，p2，…，pMIn the method, find the distance tracing point qtClosest point of trace pnearest；
According to said pnearestAnd q is as describedtCalculating a tracking position error epositionOne preferred position error is pnearestAnd q istThe Euclidean distance of;
according to said pnearestAnd q is as describedtCalculating tracking angle error eheadingOne preferred angular error is | hnearest-htL, where hnearestIs a track point pnearestCourse angle of (h)tFor unmanned vehicle current track point ptThe course angle of (d);
according to said tracking position error epositionAnd tracking angle error eheadingObtaining a tracking error etOne preferred form of calculation is et=cp
Figure 650901DEST_PATH_IMAGE002
eposition+ch
Figure 696218DEST_PATH_IMAGE002
eheadingWherein c ispAnd chIs a weight value.
According to said tracking error etCalculating the tracking ability of the unmanned vehicle (motion platform), and recording as utWherein u ist<0，utThe larger the track tracing capacity of the motion platform, the worse the track tracing capacity is, and the more the track tracing capacity is, the worse the track tracing capacity is, an optimal calculation formula is ut=-(kPet+kI∑t i=0ei+kD(et-et-1) kP, kI and kD are weights.
Step S3: updating a curvature-rate function in real time according to the tracking capability of the motion platform;
updating a curvature-rate function in real time according to the track tracking capability of the unmanned automobile;
suppose that the trajectory tracking capability of the unmanned vehicle at the current moment is ut；
If ut<ua1Then V ist(k)=Vt-1(k)+b1|ut|G1(||k-kt||)；
If ut>ua2Then V ist(k)=Vt-1(k)+b2|ut|G2(||k-kt||)；
If ua1≤ut≤ua2Then V ist(k)=Vt-1(k)；
Wherein u isa1To set a lower bound, u, of the range of trackability allowed for unmanned vehiclesa2To set an upper bound on the range of allowable traceability, b1，b2For learning rate, Vt(k) The curvature-rate function, V, obtained for the time t updatet-1(k) As a function of curvature-rate at time t-1, G1And G2Is | | | k-ktI is a function of an argument, A is a domain of the curvature-rate function, ktThe curvature of a track point closest to the current unmanned vehicle in the expected track is obtained;
the function G1Satisfies G1<0, making said updated Vt(k) Satisfy at k<0 is a non-increasing function, at k>0 is a non-decreasing function, and Vt(k)<V0(k) For any k ∈ A, a preferred G1（k，kt）=-exp{-||k-kt||2/(2sigma12) Exp is an exponential function and sigma1 is a parameter, this is preferably G1To VtSee fig. 3a for a diagram of update G1Will have poor tracking ability in the unmanned vehicle (u)t<ua1) The speed value corresponding to the curvature is properly reduced to improve the following track tracking precision;
said function G2Satisfies G2>0, making said updated Vt(k) Satisfy at k<0 is a non-increasing function, at k>0 is a non-decreasing function, and Vt(k)<V0(k) For any k ∈ A, a preferred G2（k，kt）= exp{-||k-kt||2/(2sigma22) Exp is an exponential function and sigma2 is a parameter, this is preferably G2To VtSee fig. 3b for a diagram of the update of G2Will have very good tracking ability in the unmanned automobile (u)t>ua2) The speed value corresponding to the curvature is properly increased to improve the subsequent motion efficiency.
Step S4: the motion planner of the motion platform obtains a motion planning result of the motion platform according to the updated curvature-rate function;
the motion planner of the unmanned vehicle takes the updated curvature-rate function as a constraint condition to obtain a motion planning result considering the track tracking capability;
according to the curvature-speed function V obtained by updating at the moment tt(k) Constructing a curvature-rate constraint term v ≦ V (k), wherein v is the expected rate at any time, and V (k) represents a curvature-rate function at any time;
obtaining an expected trajectory sequence and an expected speed sequence which meet the curvature-speed constraint term in a motion planner according to the curvature-speed constraint term;
the desired trajectory sequence is defined by a sequence of points { p }1，p2，…，pMDenotes wherein p isiThe information of the position, the posture, the curvature and the like of the tracing point in a space coordinate system is contained;
the desired rate sequence is composed of a rate sequence v1，v2，…，vMRepresents it.
Where M is the desired step size, i.e., the planning step size.
Step S5: the controller of the motion platform obtains a motion command according to the motion planning result and controls the motion platform to move according to the motion command;
and the controller of the unmanned automobile obtains a motion instruction according to the expected track and the expected speed result, and controls the unmanned automobile, namely the platform, to move for a time step.
And finally, repeatedly executing the step S2, learning a curvature-rate function on line according to the tracking capability evaluated in real time in the process of the movement of the unmanned automobile, and adjusting the movement planning constraint condition in real time, thereby obtaining a movement planning result giving consideration to the movement efficiency and the trajectory tracking precision.
Corresponding to the embodiment of the exercise planning method for online learning of exercise capacity, the invention also provides an embodiment of an exercise planning device for online learning of exercise capacity.
Referring to fig. 4, an exercise planning apparatus capable of learning exercise capacity online provided by an embodiment of the present invention includes one or more processors, and is configured to implement an exercise planning method capable of learning exercise capacity online in the foregoing embodiment.
The embodiment of the exercise planning device capable of learning exercise capacity on line can be applied to any equipment with data processing capacity, such as computers and other equipment or devices. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and as a logical device, the device is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for running through the processor of any device with data processing capability. From a hardware aspect, as shown in fig. 4, the present invention is a hardware structure diagram of any device with data processing capability where an exercise planning apparatus capable of learning exercise capability online is located, except for the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 4, any device with data processing capability where the apparatus is located in the embodiment may also include other hardware according to the actual function of the any device with data processing capability, which is not described again.
The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the invention. One of ordinary skill in the art can understand and implement without inventive effort.
The embodiment of the invention also provides a computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, the method for planning the exercise, in which the exercise capacity can be learned online, is implemented.
The computer readable storage medium may be an internal storage unit, such as a hard disk or a memory, of any data processing capability device described in any of the foregoing embodiments. The computer readable storage medium may also be an external storage device of the wind turbine, such as a plug-in hard disk, a Smart Media Card (SMC), an SD Card, a Flash memory Card (Flash Card), and the like, provided on the device. Further, the computer readable storage medium may include both an internal storage unit and an external storage device of any data processing capable device. The computer-readable storage medium is used for storing the computer program and other programs and data required by the arbitrary data processing-capable device, and may also be used for temporarily storing data that has been output or is to be output.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way. Although the foregoing has described the practice of the present invention in detail, it will be apparent to those skilled in the art that modifications may be made to the practice of the invention as described in the foregoing examples, or that certain features may be substituted in the practice of the invention. All changes, equivalents and modifications which come within the spirit and scope of the invention are desired to be protected.

Claims (10)

1. An exercise planning method for online learning of exercise capacity is characterized by comprising the following steps:
step S1: initializing a curvature-rate function according to the kinematic and dynamic constraints of the motion platform, wherein the curvature-rate function is as follows: a rate constraint function with curvature as an argument;
step S2: evaluating the track tracking capability of the motion platform in real time according to the actual running track and the expected track of the motion platform;
step S3: updating a curvature-rate function in real time according to the tracking capability of the motion platform;
step S4: the motion planner of the motion platform obtains a motion planning result of the motion platform according to the updated curvature-rate function;
step S5: and the controller of the motion platform obtains a motion command according to the motion planning result and controls the motion platform to move according to the motion command.
2. The exercise planning method for online learning of exercise capacity according to claim 1, wherein the step S1 specifically includes:
initializing a curvature-rate function as V according to kinematic and dynamic constraints of the motion platform0(k) = sqrt (a/| K |), a being a coefficient relating to the kinetic constraint of the motion platform, K ∈ a, a representing the curvature constraint of the motion platform, a = [ K ] ], [ K ∈ amin, Kmax]，KminAnd KmaxRespectively, the minimum and maximum curvatures of the moving platform.
3. The exercise planning method for online learning of exercise capacity according to claim 2, wherein the step S2 specifically includes:
according to the actual running track r = { q) of the moving platform1，q2，…，qN}， q1~qNFor the actual running track point, N is the recorded historical track step size, and its expected track rd = { p =1，p2，…，pM}，p1~pMFinding the actual running track point q of the motion platform at the moment of the distance t from the expected track rd for the expected running track point and M for the step length of the expected tracktClosest locus point pnearestAnd according to qtAnd pnearestCalculating the tracking error et(ii) a Then according to the tracking error etEvaluating a trajectory tracking capability u of a motion platformtWherein u ist<0，utThe larger the track tracing capacity of the motion platform, the worse the track tracing capacity is.
4. The method as claimed in claim 3, wherein the trajectory tracking capability u is a track tracking capabilitytThe expression of (c) is: u. oft=-(kPet+kI∑t i=0ei+kD(et-et-1) kP, kI and kD are weights.
5. The method as claimed in claim 3, wherein the tracking error e istThe specific expression of (A) is as follows: e.g. of the typet=cp ×eposition+ch×eheadingWherein c ispAnd chIs the weight; e.g. of the typepositionFor tracking position errors by said pnearestAnd q is as describedtCalculating the Euclidean distance; e.g. of the typeheadingTo track angular error, through the expression | hnearest-htCalculated, | hnearestIs a track point pnearestCourse angle of (h)tFor the track point p of the motion platformtThe course angle of (c).
6. The exercise planning method for online learning of exercise capacity according to claim 3, wherein the step S3 specifically includes:
if ut<ua1Then V ist(k)=Vt-1(k)+b1|ut|G1(||k-kt||)；
If ut>ua2Then V ist(k)=Vt-1(k)+b2|ut|G2(||k-kt||)；
If ua1≤ut≤ua2Then V ist(k)=Vt-1(k)；
Wherein u isa1To set the lower bound, u, of the range of trackability allowed for a moving platforma2To set an upper bound on the range of allowable trackability, b1，b2For learning rate, Vt(k) The curvature-rate function, V, obtained for the time t updatet-1(k) As a function of curvature-rate at time t-1, G1And G2Is | | | k-ktI is a function of an argument, A is a domain of the curvature-rate function, ktThe curvature of the track point closest to the motion platform at time t in the expected track is shown.
7. The method as claimed in claim 6, wherein the function G is a function of a user's ability to learn exercise on-line1Satisfies G1<0, updating the obtained Vt(k) Satisfy at k<0 is a non-increasing function, at k>0 is a non-decreasing function, and Vt(k)<V0(k) For any k ∈ A, G1（k，kt）=-exp{-||k-kt||2/(2sigma12) Exp is an exponential function, and sigma1 is a parameter;
the function G2Satisfies G2>0, making said updated Vt(k) Satisfy at k<0 is a non-increasing function, at k>0 is a non-decreasing function, and Vt(k)<V0(k) For any k ∈ A, G2（k，kt）= exp{-||k-kt||2/(2sigma22) Exp is an exponential function and sigma2 is a parameter.
8. The exercise planning method for online learning of exercise capacity according to claim 6, wherein the step S4 specifically includes:
according to the curvature-speed function V obtained by updating at the moment tt(k) Constructing a curvature-rate constraint term v ≦ V (k), wherein v is the expected rate at any time, and V (k) represents a curvature-rate function at any time;
and obtaining an expected trajectory sequence and an expected speed sequence which meet the curvature-speed constraint term, namely an action planning result, in the action planner according to the curvature-speed constraint term.
9. An exercise capacity online learning exercise planning device, comprising one or more processors for implementing an exercise capacity online learning exercise planning method according to any one of claims 1 to 8.
10. A computer-readable storage medium, having stored thereon a program which, when being executed by a processor, implements a method for exercise planning for online learning of exercise capacity according to any of claims 1 to 8.
CN202210412955.2A 2022-04-19 2022-04-19 Exercise planning method, device and medium for online learning of exercise capacity Active CN114509936B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210412955.2A CN114509936B (en) 2022-04-19 2022-04-19 Exercise planning method, device and medium for online learning of exercise capacity

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210412955.2A CN114509936B (en) 2022-04-19 2022-04-19 Exercise planning method, device and medium for online learning of exercise capacity

Publications (2)

Publication Number Publication Date
CN114509936A true CN114509936A (en) 2022-05-17
CN114509936B CN114509936B (en) 2022-09-02

Family

ID=81555340

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210412955.2A Active CN114509936B (en) 2022-04-19 2022-04-19 Exercise planning method, device and medium for online learning of exercise capacity

Country Status (1)

Country Link
CN (1) CN114509936B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106647282A (en) * 2017-01-19 2017-05-10 北京工业大学 Six-freedom-degree robot track planning method giving consideration to tail end motion error
CN110160541A (en) * 2018-08-06 2019-08-23 腾讯大地通途（北京）科技有限公司 Reconstructing method and device, storage medium, the electronic device of motion profile
CN110244735A (en) * 2019-06-24 2019-09-17 安徽农业大学 The heuristic dynamic programming control method of mobile robot tracking desired trajectory
US20200003564A1 (en) * 2018-06-27 2020-01-02 Baidu Usa Llc Reference line smoothing method using piecewise spiral curves with weighted geometry costs
CN111240345A (en) * 2020-02-11 2020-06-05 哈尔滨工程大学 Underwater robot trajectory tracking method based on double BP network reinforcement learning framework
CN113341719A (en) * 2021-06-15 2021-09-03 上海应用技术大学 AGV trajectory tracking control method integrating TEB algorithm and fuzzy control
CN113359757A (en) * 2021-06-30 2021-09-07 湖北汽车工业学院 Improved hybrid A-x algorithm unmanned vehicle path planning and trajectory tracking method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106647282A (en) * 2017-01-19 2017-05-10 北京工业大学 Six-freedom-degree robot track planning method giving consideration to tail end motion error
US20200003564A1 (en) * 2018-06-27 2020-01-02 Baidu Usa Llc Reference line smoothing method using piecewise spiral curves with weighted geometry costs
CN110160541A (en) * 2018-08-06 2019-08-23 腾讯大地通途（北京）科技有限公司 Reconstructing method and device, storage medium, the electronic device of motion profile
CN110244735A (en) * 2019-06-24 2019-09-17 安徽农业大学 The heuristic dynamic programming control method of mobile robot tracking desired trajectory
CN111240345A (en) * 2020-02-11 2020-06-05 哈尔滨工程大学 Underwater robot trajectory tracking method based on double BP network reinforcement learning framework
CN113341719A (en) * 2021-06-15 2021-09-03 上海应用技术大学 AGV trajectory tracking control method integrating TEB algorithm and fuzzy control
CN113359757A (en) * 2021-06-30 2021-09-07 湖北汽车工业学院 Improved hybrid A-x algorithm unmanned vehicle path planning and trajectory tracking method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
胡艳明等: "人在环路的机器人在线可持续示教学习", 《无人系统技术》 *

Also Published As

Publication number Publication date
CN114509936B (en) 2022-09-02

Similar Documents

Publication Publication Date Title
CN108803321B (en) Autonomous underwater vehicle track tracking control method based on deep reinforcement learning
CN106444806B (en) The drive lacking AUV three-dimensional track tracking and controlling method adjusted based on biological speed
KR20200096410A (en) Feature Extraction based on Deep Learning for LIDAR Position Estimation of Autonomous Vehicles
CN114952868B (en) 7-degree-of-freedom SRS (sounding reference Signal) type mechanical arm control method and device and piano playing robot
WO2021156516A1 (en) Multi-objective reinforcement learning using objective-specific action-value functions
CN113110504B (en) Unmanned ship path tracking method based on reinforcement learning and line-of-sight method
WO2022199146A1 (en) Robot control method and robot based on spiking neural network, and storage medium
CN107807069B (en) Self-adaptive tracking control method and system for sea surface oil spill
CN114510063B (en) Unmanned tracked vehicle and track tracking control method and system thereof
Kamran et al. Learning path tracking for real car-like mobile robots from simulation
CN114115262B (en) Multi-AUV actuator saturation cooperative formation control system and method based on azimuth information
CN113821035A (en) Unmanned ship trajectory tracking control method and device
Blekas et al. RL-based path planning for an over-actuated floating vehicle under disturbances
Zucker et al. Reinforcement planning: RL for optimal planners
Butyrev et al. Deep reinforcement learning for motion planning of mobile robots
CN113419424B (en) Modeling reinforcement learning robot control method and system for reducing overestimation
CN114509936B (en) Exercise planning method, device and medium for online learning of exercise capacity
CN114397899A (en) Bionic robot fish three-dimensional path tracking control method and device
CN117484506A (en) Robot control method, robot control device, robot and computer readable storage medium
Huang Control approach for tracking a moving target by a wheeled mobile robot with limited velocities
CN110481561A (en) Automatic driving vehicle automatic control signal generation method and device
CN114063453B (en) Helicopter system control method, system, device and medium based on reinforcement learning
CN116909136B (en) 2-DOF helicopter sliding mode control method and system based on determined learning
CN116700288A (en) Mobile robot local track planning method and system based on adaptive model predictive control
CN112130557B (en) Multi-underwater vehicle tracking control method, terminal and storage medium

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