CN115469665A - Intelligent wheelchair target tracking control method and system suitable for dynamic environment - Google Patents
Intelligent wheelchair target tracking control method and system suitable for dynamic environment Download PDFInfo
- Publication number
- CN115469665A CN115469665A CN202211127474.3A CN202211127474A CN115469665A CN 115469665 A CN115469665 A CN 115469665A CN 202211127474 A CN202211127474 A CN 202211127474A CN 115469665 A CN115469665 A CN 115469665A
- Authority
- CN
- China
- Prior art keywords
- target
- wheelchair
- intelligent wheelchair
- potential field
- obstacle
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000033001 locomotion Effects 0.000 claims abstract description 56
- 238000001514 detection method Methods 0.000 claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 230000002093 peripheral effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000005381 potential energy Methods 0.000 claims 1
- 230000006870 function Effects 0.000 description 15
- 230000007613 environmental effect Effects 0.000 description 4
- 230000004888 barrier function Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036632 reaction speed Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent wheelchair target tracking control method and system suitable for a dynamic environment, wherein the system comprises a target detection module, an obstacle detection module, a motion planning module and a motion control module; the target detection module is used for identifying a target and predicting the position of the target according to historical information; the obstacle detection module is used for identifying obstacles and predicting the probability distribution of the positions of the obstacles according to historical information; the motion planning module is used for respectively constructing a gravitational potential field and a repulsive potential field according to the predicted target position and the probability distribution of the predicted obstacle position, and the gravitational potential field and the repulsive potential field jointly act on the wheelchair to generate a safe path; and the motion control module is used for tracking the safety path generated by the motion planning module and executing the motion instruction of the intelligent wheelchair. The invention solves the problem that the current intelligent wheelchair is difficult to realize the autonomous following dynamic target in the dynamic obstacle environment.
Description
Technical Field
The invention relates to the technical field of intelligent wheelchair target tracking, in particular to an intelligent wheelchair target tracking control method and system suitable for a dynamic environment.
Background
The social aging degree is more and more serious, and the intelligent wheelchair has extremely wide development prospect and social value as a special tool capable of improving the living quality and the freedom of movement of the old and the disabled. Some advanced intelligent wheelchairs at present have a certain navigation obstacle avoidance function, but still are difficult to deal with complex environments, and complete complex tasks, such as following dynamic targets under a dynamic environment, and needing to consider the motion trend of following objects and avoiding objects, and the motion track of the traditional intelligent wheelchairs based on static scene planning is difficult to deal with the dynamic environment which is changeable instantly.
Disclosure of Invention
Aiming at the defects, the invention provides an intelligent wheelchair target tracking control method and system suitable for a dynamic environment, and aims to solve the problem that the current intelligent wheelchair is difficult to automatically follow a dynamic target under the condition of a dynamic obstacle.
In order to achieve the purpose, the invention adopts the following technical scheme:
an intelligent wheelchair target tracking control method suitable for a dynamic environment comprises the following steps:
step S1: identifying a target and predicting a target location from historical information;
step S2: identifying an obstacle and predicting a probability distribution of the obstacle position from the historical information;
and step S3: respectively constructing an attraction force potential field and a repulsion force potential field according to the predicted target position and the predicted probability distribution of the position of the obstacle, and generating a safe path of the intelligent wheelchair under the combined action of the attraction force potential field and the repulsion force potential field;
and step S4: and generating an intelligent wheelchair motion instruction based on the intelligent wheelchair safety path controller, and executing the intelligent wheelchair motion instruction.
Preferably, the step S1 specifically includes the following steps:
step S11: acquiring image data of a target, selecting a Haar-like feature as a target gray feature, processing the image data by using a semi-supervised On-line Boosting algorithm, calculating to obtain an image space position coordinate of the target, and further obtaining a task space position coordinate of the target based On depth image information;
step S12: predicting the task space pose of the target by using a Least square Support Vector regression (LS-SVR) algorithm based on the historical data of the target pose coordinate in the task space to obtain predicted dynamic target pose information P p (x p ,y p )；
Step S13: according to the predicted dynamic object pose information P p (x p ,y p ) And a set following orientation theta f Calculating the final desired position P of the wheelchair D (x D ,y D )。
Preferably, in step S11, the On-line Boosting algorithm specifically includes the following steps:
step S111: selecting a target area, and amplifying the target area according to a certain proportion to form a search area; the search area comprises a possible position where the target is located after the target moves next time; after the target area and the search area are determined, selecting samples according to the target position to train a prior classifier and a tracker in sequence, wherein 5 samples are used for each training, and the 5 samples comprise the target and 4 samples around the target;
step S112: dividing the search area into a plurality of small blocks, wherein the size of the divided small blocks is the same as that of the target area, and the divided small blocks have overlapping parts; evaluating all the divided small blocks by using the tracker trained in the previous stage, calculating the credibility of each small block as a target, and selecting the small block with the highest credibility as the position of the target at the next moment;
step S113: and according to the predicted new target position, reselecting a sample by using a tracker for training, and updating the parameters of the classifier, wherein the label of the sample is determined by using the new target position and 4 peripheral small blocks as samples through a prior classifier.
Preferably, in step S12, the LS-SVR algorithm specifically includes the following steps:
step S121: given data set
Selecting the appropriate model parameter gamma>0; wherein d is the training set size and m is the test set size; l is the data set size; x is the number of i Line i, y representing the input sample i Representing the respective output value;
step S122: selecting radial basis functions
As a kernel function;
step S123: calculation of p = H -1 y，q＝H -1 L and s = L T q; wherein p is the wheelchair pose, H is the positive definite matrix, q is the transformation matrix, L = (1, ·, 1) T ∈R l ；
Step S124: calculation of b * ＝η T y/s and a * = p-bq; wherein b is an offset vector; eta is a parameter matrix; a is * A vector composed of Lagrange multipliers;
step S125: constructing a regression function
Preferably, in step S13, the final desired position of the wheelchair is obtained from equation (1):
step S13: the predicted target task space pose is P p (x p ,y p ) Set following azimuth θ f According to the formula (1), the expected position P of the intelligent wheelchair can be obtained D (x D ,y D )：
Where d is the relative distance desired to be maintained in following.
Preferably, the step S2 specifically includes the following steps:
step S21: passing the position of the ith obstacle at the last moment
And the position of the obstacle at the current time
Calculating to obtain the position of the obstacle at the next moment
Step S22: the displacement of the obstacle at two adjacent times is recorded as (o) x,i ,o y,i ) Performing probability analysis on the obstacle to obtain the expected value mu of the obstacle x And mu y Variance σ x And σ y Covariance σ xy And correlation coefficient ρ xy ：
N is the historical data volume participating in prediction, and the latest N historical sequences are intercepted as predictions, so that data expansion in the later planning stage can be avoided;
step S23: dividing the coordinate system into a number of grids, the probability density U of each grid (m, n) ob (m, n) can be obtained by the following formula:
wherein, U ob (m, n) is the probability density of each grid (m, n).
Preferably, the step S3 specifically includes the following steps:
step S31: according to the calculation principle of the step S13, the final expected position P of the intelligent wheelchair is obtained D (x D ,y D ) And forming a gravitational potential field as a target point according to the final expected position of the intelligent wheelchair, wherein the gravitational potential U generated by the gravitational potential field at the grid position (m, n) att (m, n) is represented by formula (8):
where S represents the size of the actual motion environment, (x) D ,y D ) Representing a target location;
step S32: calculating the resultant force field U of the current point (x, y) according to the superposition of the attraction force field and the repulsion force field t (m, n); the specific calculation is as follows:
U t (m,n)＝U att (m,n)+U ob (m,n) (9)
step S33: and carrying out deviation calculation on the generated potential field to obtain the following potential field gradient:
and (3) calculating the expected position of the intelligent wheelchair at the next moment according to the potential field gradient, wherein the specific calculation is as follows:
wherein x is d And y d Representing the expected X-axis coordinate and Y-axis coordinate of the wheelchair at the next moment; theta d Representing the expected posture of the wheelchair at the next moment, and X and Y represent the X-axis coordinate and the Y-axis coordinate of the wheelchair at the current moment; d represents the moving distance of the wheelchair in one period and can be regarded as a reference speed value, and the smaller the value is, the safer the movement of the intelligent wheelchair is; and R is a reference norm value of the movement distance of the intelligent wheelchair in one period.
R represents the following:
preferably, in step S4, the generation of the motion command of the intelligent wheelchair is a trajectory planned by a potential field method tracked by an inversion controller based on a kinematic model, and the method specifically includes the following steps:
step S41: introducing virtual input alpha, and taking the alpha according to the kinematic equation of the robot
Wherein v is the linear velocity of the robot, and the Lyapunov function is used for judging the stability of the nonlinear system; let Lyapunov function V 1 Is composed of
Wherein e x Indicating the position error of the wheelchair in the X direction, e y Indicating a position error of the wheelchair in the Y direction;
is obtained by the formula (16)
By designing the virtual quantity alpha such that
Wherein x is d And y d Representing the desired X-axis and Y-axis coordinates of the wheelchair at the next moment, then
Wherein, c 1 、c 2 Is an adjustable parameter;
order to
The linear velocity v and the virtual control law α are designed as follows:
ensuring that equation (20) holds;
step S42: let e = α - θ, define the Lyapunov function V 2 Comprises the following steps:
then
The angular velocity control law ω is designed as:
wherein c is 3 To adjust the parameters, then
Wherein C is m Is a constant number, C m ≤min(c 1 ,c 2 ,c 3 )；
Then
I.e. V 2 (t) converges exponentially to zero, so that t → ∞ time, e x →0，e y →0，θ→θ d And converges exponentially.
Another aspect of the application provides an intelligent wheelchair target tracking control system adapted to a dynamic environment, the system comprising a target detection module, an obstacle detection module, a motion planning module and a motion control module;
the target detection module is used for identifying a target and predicting the position of the target according to historical information;
the obstacle detection module is used for identifying obstacles and predicting probability distribution of obstacle positions according to historical information;
the motion planning module is used for respectively constructing an attraction potential field and a repulsion potential field according to the predicted target position and the probability distribution of the predicted obstacle position, and the attraction potential field and the repulsion potential field jointly act to generate a safety path of the intelligent wheelchair;
the motion control module is used for tracking the safety path generated by the motion planning module and executing the motion instruction of the intelligent wheelchair.
The technical scheme provided by the embodiment of the application can have the following beneficial effects:
in the scheme, a target is identified by combining a target detection module with a semi-supervised On-lineBoosting algorithm, and the predicted target position is analyzed by adopting a least square support vector regression method according to historical data; meanwhile, the obstacle detection module identifies the obstacles, analyzes the probability distribution of the predicted obstacle positions according to historical data, generates a gravitational potential field based on the predicted target positions, generates a repulsive potential field based on the probability distribution of the predicted obstacle positions, and plans a safe following path through the combined action of the two potential fields. The intelligent wheelchair moves according to the planned following path, and stable, safe and efficient following of the target object in a dynamic environment by the intelligent wheelchair is achieved.
Drawings
FIG. 1 is a functional block diagram of an intelligent wheelchair target tracking control system adapted to a dynamic environment;
fig. 2 is a hardware architecture framework diagram of the intelligent wheelchair.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
An intelligent wheelchair target tracking control method suitable for a dynamic environment comprises the following steps:
step S1: identifying a target and predicting a target location from historical information;
step S2: identifying an obstacle and predicting a probability distribution of the obstacle position from the historical information;
and step S3: respectively constructing a gravitational potential field and a repulsive potential field according to the predicted target position and the predicted probability distribution of the position of the obstacle, and generating a safe path of the intelligent wheelchair under the combined action of the gravitational potential field and the repulsive potential field;
and step S4: and generating an intelligent wheelchair motion instruction based on the intelligent wheelchair safety path controller, and executing the intelligent wheelchair motion instruction.
The method comprises the steps of identifying a target through a target detection module, namely a visual sensor in combination with a semi-supervised On-line Boosting algorithm, and analyzing a predicted target position by adopting a least square support vector regression method according to historical data; meanwhile, an obstacle detection module, namely a laser sensor, identifies obstacles, analyzes the probability distribution of the predicted obstacle positions according to historical data, generates a gravitational potential field based on the predicted target positions, generates a repulsive potential field based on the probability distribution of the predicted obstacle positions, and plans a safe following path through the combined action of the two potential fields. The intelligent wheelchair moves according to the planned following path, and stable, safe and efficient following of the target object by the intelligent wheelchair in a dynamic environment is achieved.
Preferably, the step S1 specifically includes the following steps:
step S11: acquiring image data of a target, selecting a Haar-like feature as a target gray feature, processing the image data by using a semi-supervised On-line Boosting algorithm, calculating to obtain an image space position coordinate of the target, and further obtaining a task space position coordinate of the target based On depth image information;
step S12: predicting the task space pose of the target by using a Least square Support Vector regression (LS-SVR) algorithm based on the historical data of the target pose coordinate in the task space to obtain predicted dynamic target pose information P p (x p ,y p )；
Step S13: from predicted dynamic object pose information P p (x p ,y p ) And a set following orientation theta f Calculating the final desired position P of the wheelchair D (x D ,y D )。
In the embodiment, image data is processed through a semi-supervised On-line Boosting algorithm, the position coordinates of the target in the image space are obtained through calculation, and further the position coordinates of the target point in the task space are obtained based On the depth image information. And then, predicting the target position by using an LS-SVR algorithm based on the historical data of the target position coordinates to obtain the reliable prediction of the dynamic target. Therefore, the time lag of the system is reduced, and the reaction speed of the system following the dynamic target is improved.
Preferably, in step S11, the On-line Boosting algorithm specifically includes the following steps:
step S111: selecting a target area, and amplifying the target area according to a certain proportion to form a search area; the search area comprises a possible position where the target is located after moving next time; after the target area and the search area are determined, selecting samples to train a prior classifier and a tracker in sequence according to the position of the target, wherein 5 samples are used for each training, and the 5 samples comprise the target and 4 samples around the target;
step S112: dividing the search area into a plurality of small blocks, wherein the size of the divided small blocks is the same as that of the target area, and the divided small blocks have overlapping parts; evaluating all the divided small blocks by using the tracker trained in the previous stage, calculating the credibility of each small block as a target, and selecting the small block with the highest credibility as the position of the target at the next moment;
step S113: and according to the predicted new target position, reselecting a sample by using a tracker for training, and updating the parameters of the classifier, wherein the label of the sample is determined by using the new target position and 4 peripheral small blocks as samples through a prior classifier.
Specifically, the semi-supervised On-line Boosting algorithm considers that the labels of all samples in the update phase are unknown, except that the labels of the samples in the initial phase are certain. The labels of the samples in the update phase are determined by an offline a priori classifier that only carries the original target information. The prior classifier is only trained in the following target initialization phase and does not participate in updating thereafter. It therefore contains only the information of the original object, i.e. a priori information.
There are two strong classifiers, a priori classifier and a tracker, in the semi-supervised On-line Boosting algorithm. Wherein, the prior classifier does not participate in updating, and is only used for evaluating the matching degree between the current sample and the initial target sample. And according to the matching degree between the samples, giving a label to the current sample. When the exemplar labels are assigned, the exemplars may update the tracker in a supervised learning manner. The whole semi-supervised On-lineBoosting algorithm mainly comprises an initialization training stage and a tracking stage, wherein the tracking stage is divided into a prediction stage and an updating stage.
Preferably, in step S12, the LS-SVR algorithm specifically includes the following steps:
step S121: given data set
Selecting the appropriate model parameter gamma>0; wherein d is the training set size and m is the test set size; l is the data set size; x is the number of i Line i, y representing the input sample i Representing the respective output value;
step S122: selecting radial basis functions
As a kernel function;
step S123: calculation of p = H -1 y，q＝H -1 L and s = L T q; wherein p is the wheelchair pose, H is the positive definite matrix, q is the transformation matrix, L = (1, ·, 1) T ∈R l ；
Step S124: calculation of b * ＝η T y/s and a * = p-bq; wherein b is an offset vector; eta is a parameter matrix; a is * A vector composed of Lagrange multipliers;
step S125: constructing a regression function
The problem of short sight generally exists when the intelligent wheelchair tracks the target, only the current position is considered, and the position of the target appearing in the future is not considered. Therefore, in order to better track the dynamic target, it is necessary to know the motion law of the dynamic target in the environment, i.e. to predict the dynamic target. In this embodiment, an LS-SVR (least square support vector regression) method is adopted to consider dynamic target trajectory prediction as a problem of analysis and prediction of a small sample and nonlinear time series data, where time is input and dynamic target trajectory corresponding to the time is output. And learning a motion model of the target, namely a nonlinear relation between time and the dynamic target according to the track in a past period of time, so as to predict the motion track of the target in a future period of time.
In particular, the nonlinear relationship can be well approximated by the LS-SVR, so that the trajectory of the target can be predicted. When the SVR is used for track prediction, the problem is regarded as a prediction problem of time series data, namely, a mapping relation between time and a dynamic target track is established. Input variable X = (t) in training data 1 ,t 2 ,…,t n ) Is a time series, corresponding output variable Y = (p) 1 ,p 2 ,…,p n ) Wherein p is k Representing the pose at time k.
Different kernel functions are selected to greatly influence the regression effect of the SVR, the motion trajectory of the dynamic target is generally nonlinear, and in the embodiment, a Radial Basis Function (RBF) kernel function is adopted, and by simply processing data, a better effect can be achieved on both linear regression and nonlinear regression.
Preferably, in step S13, the final desired position of the wheelchair can be obtained by equation (1):
step S13: the predicted target task space pose is P p (x p ,y p ) Set following azimuth θ f According to the formula (1), the expected position P of the intelligent wheelchair can be obtained D (x D ,y D )：
Where d is the relative distance desired to be maintained in following.
In particular, the desired position P of the intelligent wheelchair D (x D ,y D ) The calculation of (2) enables the intelligent wheelchair to better plan a walking path, and further realizes that the intelligent wheelchair autonomously follows a dynamic target object under the condition of a dynamic barrier.
Preferably, the step S2 specifically includes the following steps:
step S21: passing the position of the ith obstacle at the last moment
And the position of the obstacle at the current time
Calculating to obtain the position of the obstacle at the next moment
Step S22: the displacement of the obstacle at two adjacent times is recorded as (o) x,i ,o y,i ) Performing probability analysis on the obstacle to obtain the expected value mu of the obstacle x And mu y Variance σ x And σ y Covariance σ xy And correlation coefficient ρ xy ：
N is the historical data volume participating in prediction, and the latest N historical sequences are intercepted as predictions, so that data expansion in the later planning stage can be avoided;
step S23: dividing the coordinate system into a number of grids, each grid (m, n) having a probability density U ob (m, n) can be obtained by the following formula:
wherein, U ob (m, n) is the probability density of each grid (m, n).
Specifically, when the intelligent wheelchair tracks a target point in a dynamic environment, effective navigation and obstacle avoidance are indispensable parts, and the intelligent wheelchair needs to plan a collision-free safe path. According to the method, environmental information around the intelligent wheelchair is collected on the intelligent wheelchair through a laser sensor and transmitted to an industrial personal computer, the environmental information is connected and communicated through Ethernet, and an application program processes the collected data, predicts Probability distribution of the position of the obstacle at the next moment and generates a collision-free track.
Preferably, the step S3 specifically includes the following steps:
step S31: according to the calculation principle of the step S13, the final expected position P of the intelligent wheelchair is obtained D (x D ,y D ) And forming a gravitational potential field as a target point according to the final expected position of the intelligent wheelchair, wherein the gravitational potential U generated by the gravitational potential field at the grid position (m, n) att (m, n) is represented by formula (8):
where S represents the size of the actual motion environment, (x) D ,y D ) Representing a target location;
step S32: calculating the resultant force field U of the current point (x, y) according to the superposition of the attractive force potential field and the repulsive force potential field t (m, n); the specific calculation is as follows:
U t (m,n)＝U att (m,n)+U ob (m,n) (9)
step S33: and carrying out deviation calculation on the generated potential field to obtain the following potential field gradient:
and (3) calculating the expected position of the intelligent wheelchair at the next moment according to the potential field gradient, wherein the specific calculation is as follows:
wherein x is d And y d Representing the expected X-axis coordinate and Y-axis coordinate of the wheelchair at the next moment; theta.theta. d Representing the expected posture of the wheelchair at the next moment, and X and Y represent the X-axis coordinate and the Y-axis coordinate of the wheelchair at the current moment; d represents the moving distance of the wheelchair in one period and can be regarded as a reference speed value, and the smaller the value is, the safer the movement of the intelligent wheelchair is; and R is a reference norm value of the movement distance of the intelligent wheelchair in a period.
R represents the following:
specifically, a collision-free track is planned for the intelligent wheelchair through a random potential field method, so that the intelligent wheelchair can safely and effectively track a dynamic target object.
Preferably, in step S4, the generation of the motion command of the intelligent wheelchair is a trajectory planned by a potential field method tracked by an inversion controller based on a kinematic model, and the method specifically includes the following steps:
step S41: the introduction of virtual input alpha is obtained according to the kinematic equation of the robot
Wherein v is the linear velocity of the robot, and the Lyapunov function is used for judging the stability of the nonlinear system; let Lyapunov function V 1 Is composed of
Wherein e x Indicating the position error of the wheelchair in the X direction, e y Indicating a position error of the wheelchair in the Y direction;
e x ＝x d -x (15)
e y ＝y d -y
is obtained by the formula (16)
By designing the virtual quantity alpha such that
Wherein x is d And y d Indicating wheelThe expected X-axis coordinate and Y-axis coordinate at the next moment of the chair, then
Wherein, c 1 、c 2 Is an adjustable parameter;
order to
The linear velocity v and the virtual control law α are designed as follows:
ensuring that equation (20) holds;
step S42: let e = α - θ, define the Lyapunov function V 2 Comprises the following steps:
then
The angular velocity control law ω is designed as:
wherein c is 3 To adjust the parameters, then
Wherein C is m Is a constant number, C m ≤min(c 1 ,c 2 ,c 3 )；
Then
I.e. V 2 (t) converges exponentially to zero, so that t → ∞ time, e x →0，e y →0，θ→θ d And converges exponentially.
The trajectory planned by the potential field method is tracked by adopting an inversion controller based on a kinematic model. And in the tracking process, the controller positions the intelligent wheelchair in real time based on the odometer of the incremental coding disc and provides motion feedback for the controller.
In one embodiment, in step S41, let x e ＝0,y e =0, then
To achieve theta tracking d Step S42 is to ensure that θ tracks α.
Another aspect of the application provides an intelligent wheelchair target tracking control system adapted to a dynamic environment, the system comprising a target detection module, an obstacle detection module, a motion planning module and a motion control module;
the target detection module is used for identifying a target and predicting the position of the target according to historical information;
the obstacle detection module is to identify an obstacle and predict a probability distribution of obstacle locations;
the motion planning module is used for respectively constructing an attraction potential field and a repulsion potential field according to the predicted target position and the probability distribution of the predicted obstacle position, and the attraction potential field and the repulsion potential field jointly act to generate a safety path of the intelligent wheelchair;
the motion control module is used for tracking the safety path generated by the motion planning module and executing the motion instruction of the intelligent wheelchair.
In the scheme, the intelligent wheelchair target tracking control system is suitable for a dynamic environment, and as shown in fig. 1-2, the intelligent wheelchair comprises a power supply assembly, a main control assembly, a motion assembly and a sensing assembly; and a power supply assembly consisting of a 24V direct-current power supply and a transformer provides a direct-current power supply for the whole intelligent wheelchair system. The industrial personal computer is used as a main control assembly of the intelligent wheelchair system and carries a sensing assembly consisting of a visual sensor and a laser sensor. The wheel is connected with a direct current motor provided with a coding disc and a servo driver to form a motion assembly of the intelligent wheelchair. When the intelligent wheelchair performs follow-up control on a target, the sensing assembly acquires target information and environment information, sends the target information and the environment information to the industrial personal computer for data processing and analysis, and the main control assembly sends an instruction to the movement assembly to control the rotating speed and the direction of the wheel.
Further, in the power supply assembly, a 24V direct current power supply directly supplies power to the industrial personal computer, the servo driver and the laser sensor, and the display screen is connected with the transformer through the 24V direct current power supply and is stepped down to 12V for supplying power. In the main control assembly, on one hand, the industrial personal computer receives and analyzes data of the external sensor by utilizing a written application program; on the other hand, control instructions are sent to the intelligent wheelchair through the application program to execute different tasks. In the sensing assembly, a laser sensor and an industrial personal computer are connected and communicated through an Ethernet port, and the laser sensor acquires environmental information around the intelligent wheelchair and transmits the environmental information to the industrial personal computer; the visual sensor is connected with the industrial personal computer through a USB, and transmits acquired image information to the industrial personal computer. In the motion assembly, a servo driver is connected with an industrial personal computer through a CAN bus so as to receive a control command and send related data of motion of the intelligent wheelchair, and the driver receives the control command, generates the rotating speed of a motor through the processing and conversion of a built-in integrated chip and transmits the rotating speed to the motor to drive the intelligent wheelchair to move; meanwhile, a data interface arranged in the driver can acquire the accurate position of the motor in real time through an incremental encoder and reversely transmit data to the industrial personal computer, so that feedback control of state information is realized.
According to the scheme, the target is identified through the target detection module, the target position is predicted according to historical information, meanwhile, the obstacle detection module identifies the obstacle, the probability distribution of the obstacle position is predicted according to the historical information, the attraction potential field is generated based on the predicted target position, the repulsion potential field is generated based on the probability distribution of the predicted obstacle position, and a safe following path is planned through the combined action of the two potential fields. The method is different from the traditional target following method, and based on the historical data of the target detection module and the barrier detection module, the system has the prediction capability on the movement of the target and the barrier, so that the planned track has better adaptability to the dynamic environment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and not to be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made in the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (9)
1. An intelligent wheelchair target tracking control method suitable for a dynamic environment is characterized by comprising the following steps: the method comprises the following steps:
step S1: identifying a target and predicting a target location from historical information;
step S2: identifying an obstacle and predicting a probability distribution of the obstacle position from the historical information;
and step S3: respectively constructing an attraction force potential field and a repulsion force potential field according to the predicted target position and the predicted probability distribution of the position of the obstacle, and generating a safe path of the intelligent wheelchair under the combined action of the attraction force potential field and the repulsion force potential field;
and step S4: and generating an intelligent wheelchair motion instruction based on the intelligent wheelchair safety path controller, and executing the intelligent wheelchair motion instruction.
2. The intelligent wheelchair target tracking control method adapted to the dynamic environment as claimed in claim 1, characterized in that: in step S1, the method specifically includes the following steps:
step S11: acquiring image data of a target, selecting a Haar-like feature as a target gray feature, processing the image data by using a semi-supervised On-line Boosting algorithm, calculating to obtain an image space position coordinate of the target, and further obtaining a task space position coordinate of the target based On depth image information;
step S12: predicting the task space pose of the target by using a Least square support vector regression (LS-SVR) algorithm based on the historical data of the target pose coordinate in the task space to obtain predicted dynamic target pose information P p (x p ,y p )；
Step S13: from predicted dynamic object pose information P p (x p ,y p ) And a set following orientation theta f Calculating the final desired position P of the wheelchair D (x D ,y D )。
3. The intelligent wheelchair target tracking control method adapted to the dynamic environment as claimed in claim 2, characterized in that: in step S11, the On-lineBoosting algorithm specifically includes the following steps:
step S111: selecting a target area, and amplifying the target area according to a certain proportion to form a search area; the search area comprises a possible position where the target is located after the target moves next time; after the target area and the search area are determined, selecting samples to train a prior classifier and a tracker in sequence according to the position of the target, wherein 5 samples are used for each training, and the 5 samples comprise the target and 4 samples around the target;
step S112: dividing the search area into a plurality of small blocks, wherein the size of the divided small blocks is the same as that of the target area, and the divided small blocks have overlapping parts; evaluating all the divided small blocks by using the tracker trained in the previous stage, calculating the credibility of each small block as a target, and selecting the small block with the highest credibility as the position of the target at the next moment;
step S113: and according to the predicted new target position, reselecting a sample by using a tracker for training, and updating the parameters of the classifier, wherein the label of the sample is determined by using the new target position and 4 peripheral small blocks as samples through a prior classifier.
4. The intelligent wheelchair target tracking control method adapted to the dynamic environment as claimed in claim 2, characterized in that: in step S12, the LS-SVR algorithm specifically includes the following steps:
step S121: given data set
Selecting the appropriate model parameter gamma>0; wherein d is the training set size and m is the test set size; l is the data set size; x is the number of i Line i, y representing the input sample i Representing the respective output value;
step S122: selecting radial basis functions
As a kernel function;
step S123: calculation of p = H -1 y，q＝H -1 L and s = L T q; wherein p is the pose of the wheelchair, H is the positive definite matrix, q is the transformation matrix, L = (1, · ·, 1) T ∈R l ；
Step S124: calculation of b * ＝η T y/s and a * = p-bq; wherein b is an offset vector; eta is a parameter matrix; a is * A vector composed of Lagrange multipliers;
step S125: constructing a regression function
5. The intelligent wheelchair target tracking control method adapted to dynamic environment as claimed in claim 2, characterized in that: in step S13, the final desired position of the wheelchair can be obtained from equation (1):
step S13: the predicted target task space pose is P p (x p ,y p ) Set following azimuth θ f According to the formula (1), the expected position P of the intelligent wheelchair can be obtained D (x D ,y D )：
Where d is the relative distance desired to be maintained in following.
6. The intelligent wheelchair target tracking control method adapted to the dynamic environment as claimed in claim 1, characterized in that: in step S2, the method specifically includes the following steps:
step S21: passing the position of the ith obstacle at the last moment
And the position of the obstacle at the current time
Calculating to obtain the position of the obstacle at the next moment
Step S22: the displacement of the obstacle at two adjacent times is recorded as (o) x,i ,o y,i ) Performing probabilistic analysis on the obstacle to obtain the obstacleExpected value mu x And mu y Variance σ x And σ y Covariance σ xy And correlation coefficient ρ xy ：
N is the historical data volume participating in prediction, and the latest N historical sequences are intercepted as predictions, so that data expansion in the later planning stage can be avoided;
step S23: dividing the coordinate system into a number of grids, the probability density U of each grid (m, n) ob (m, n) can be obtained by the following formula:
wherein, U ob (m, n) is the probability density of each grid (m, n).
7. The intelligent wheelchair target tracking control method adapted to dynamic environment as claimed in claim 2, characterized in that: in step S3, the method specifically includes the following steps:
step S31: according to the calculation principle of the step S13, the final expected position P of the intelligent wheelchair is obtained D (x D ,y D ) And according to the final expectation of the intelligent wheelchairThe position as a target point forms an attractive potential field, wherein the potential energy U generated by the attractive potential field at the grid position (m, n) att (m, n) is represented by formula (8):
where S represents the size of the actual motion environment, (x) D ,y D ) Representing a target location;
step S32: calculating the resultant force field U of the current point (x, y) according to the superposition of the attractive force potential field and the repulsive force potential field t (m, n); the specific calculation is as follows:
U t (m,n)＝U att (m,n)+U ob (m,n) (9)
step S33: and carrying out deviation calculation on the generated potential field to obtain the following potential field gradient:
and (3) calculating the expected position of the intelligent wheelchair at the next moment according to the potential field gradient, wherein the specific calculation is as follows:
wherein x is d And y d Representing the expected X-axis coordinate and Y-axis coordinate of the wheelchair at the next moment; theta d Representing the expected posture of the wheelchair at the next moment, and X and Y representing the X-axis coordinate and the Y-axis coordinate of the wheelchair at the current moment; d represents the moving distance of the wheelchair in one period and can be regarded as a reference speed value, and the smaller the value is, the safer the movement of the intelligent wheelchair is; and R is a reference norm value of the movement distance of the intelligent wheelchair in one period.
R represents the following:
8. the intelligent wheelchair target tracking control method adapted to dynamic environment of claim 1, characterized in that: in step S4, the generation of the motion instruction of the intelligent wheelchair is a potential field method planned track tracked by an inversion controller based on a kinematic model, and specifically includes the following steps:
step S41: introducing virtual input alpha, and taking the alpha according to the kinematic equation of the robot
Wherein v is the linear velocity of the robot, and the Lyapunov function is used for judging the stability of the nonlinear system; let Lyapunov function V 1 Is composed of
Wherein e x Indicating the position error of the wheelchair in the X direction, e y Indicating a position error of the wheelchair in the Y direction;
is obtained by the formula (16)
By designing the virtual quantity alpha such that
Wherein x is d And y d Representing the desired X-axis and Y-axis coordinates of the wheelchair at the next moment, then
Wherein, c 1 、c 2 Is an adjustable parameter;
order to
The linear velocity v and the virtual control law α are designed as follows:
ensuring that equation (20) holds;
step S42: let e = α - θ, define the Lyapunov function V 2 Comprises the following steps:
then
Wherein, ω is the angular velocity of the robot, and the angular velocity control law ω is designed as:
wherein c is 3 To adjust the parameters, then
Wherein C is m Is a constant number, C m ≤min(c 1 ,c 2 ,c 3 )；
Then
I.e. V 2 (t) converges exponentially to zero, so that t → ∞ time, e x →0，e y →0，θ→θ d And converges exponentially.
9. The utility model provides an intelligence wheelchair target tracking control system who is adapted to dynamic environment which characterized in that: the intelligent wheelchair target tracking control method adapted to a dynamic environment of any one of claims 1-8, the system comprising a target detection module, an obstacle detection module, a motion planning module and a motion control module;
the target detection module is used for identifying a target and predicting the position of the target according to historical information;
the obstacle detection module is used for identifying obstacles and predicting probability distribution of obstacle positions according to historical information;
the motion planning module is used for respectively constructing an attraction potential field and a repulsion potential field according to the predicted target position and the probability distribution of the predicted obstacle position, and the attraction potential field and the repulsion potential field jointly act to generate a safety path of the intelligent wheelchair;
the motion control module is used for tracking the safety path generated by the motion planning module and executing the motion instruction of the intelligent wheelchair.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211127474.3A CN115469665B (en) | 2022-09-16 | 2022-09-16 | Intelligent wheelchair target tracking control method and system suitable for dynamic environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211127474.3A CN115469665B (en) | 2022-09-16 | 2022-09-16 | Intelligent wheelchair target tracking control method and system suitable for dynamic environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115469665A true CN115469665A (en) | 2022-12-13 |
| CN115469665B CN115469665B (en) | 2023-07-04 |
Family
ID=84332962
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211127474.3A Active CN115469665B (en) | 2022-09-16 | 2022-09-16 | Intelligent wheelchair target tracking control method and system suitable for dynamic environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115469665B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2251157A1 (en) * | 2009-05-15 | 2010-11-17 | Honda Research Institute Europe GmbH | Autonomous robots with planning in unpredictable, dynamic and complex environments |
| CN108673501A (en) * | 2018-05-17 | 2018-10-19 | 中国科学院深圳先进技术研究院 | A kind of the target follower method and device of robot |
| CN110111880A (en) * | 2019-04-22 | 2019-08-09 | 北京航空航天大学 | The Artificial Potential Field paths planning method and device based on obstacle classification of flexible needle |
| CN110471422A (en) * | 2019-08-29 | 2019-11-19 | 南京理工大学 | The detection of obstacles and automatic obstacle avoiding method of intelligent wheel chair |
| CN111258824A (en) * | 2020-01-18 | 2020-06-09 | 重庆邮电大学 | Increment check point fault tolerance method based on artificial potential field in cloud computing |
| CN112344943A (en) * | 2020-11-20 | 2021-02-09 | 安徽工程大学 | Intelligent vehicle path planning method for improving artificial potential field algorithm |
| CN112683288A (en) * | 2020-11-30 | 2021-04-20 | 北方工业大学 | Intelligent guide robot system and method for assisting blind in crossing street in intersection environment |
| CN113342047A (en) * | 2021-06-23 | 2021-09-03 | 大连大学 | Unmanned aerial vehicle path planning method for improving artificial potential field method based on obstacle position prediction in unknown environment |
| CN113759936A (en) * | 2021-09-30 | 2021-12-07 | 哈尔滨工业大学 | Path planning method combining proportion guidance method and artificial potential field method suitable for dynamic target tracking |
| CN114063621A (en) * | 2021-11-22 | 2022-02-18 | 燕山大学 | Wheeled robot formation tracking and obstacle avoidance control method |
| CN114442637A (en) * | 2022-02-10 | 2022-05-06 | 北京理工大学 | Unmanned vehicle local dynamic obstacle avoidance path planning method |
| CN114821764A (en) * | 2022-01-25 | 2022-07-29 | 哈尔滨工程大学 | Gesture image recognition method and system based on KCF tracking detection |
| CN114848315A (en) * | 2022-05-05 | 2022-08-05 | 广东工业大学 | Intelligent wheelchair man-machine cooperative control system based on surface electromyogram signals |
-
2022
- 2022-09-16 CN CN202211127474.3A patent/CN115469665B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2251157A1 (en) * | 2009-05-15 | 2010-11-17 | Honda Research Institute Europe GmbH | Autonomous robots with planning in unpredictable, dynamic and complex environments |
| CN108673501A (en) * | 2018-05-17 | 2018-10-19 | 中国科学院深圳先进技术研究院 | A kind of the target follower method and device of robot |
| CN110111880A (en) * | 2019-04-22 | 2019-08-09 | 北京航空航天大学 | The Artificial Potential Field paths planning method and device based on obstacle classification of flexible needle |
| CN110471422A (en) * | 2019-08-29 | 2019-11-19 | 南京理工大学 | The detection of obstacles and automatic obstacle avoiding method of intelligent wheel chair |
| CN111258824A (en) * | 2020-01-18 | 2020-06-09 | 重庆邮电大学 | Increment check point fault tolerance method based on artificial potential field in cloud computing |
| CN112344943A (en) * | 2020-11-20 | 2021-02-09 | 安徽工程大学 | Intelligent vehicle path planning method for improving artificial potential field algorithm |
| CN112683288A (en) * | 2020-11-30 | 2021-04-20 | 北方工业大学 | Intelligent guide robot system and method for assisting blind in crossing street in intersection environment |
| CN113342047A (en) * | 2021-06-23 | 2021-09-03 | 大连大学 | Unmanned aerial vehicle path planning method for improving artificial potential field method based on obstacle position prediction in unknown environment |
| CN113759936A (en) * | 2021-09-30 | 2021-12-07 | 哈尔滨工业大学 | Path planning method combining proportion guidance method and artificial potential field method suitable for dynamic target tracking |
| CN114063621A (en) * | 2021-11-22 | 2022-02-18 | 燕山大学 | Wheeled robot formation tracking and obstacle avoidance control method |
| CN114821764A (en) * | 2022-01-25 | 2022-07-29 | 哈尔滨工程大学 | Gesture image recognition method and system based on KCF tracking detection |
| CN114442637A (en) * | 2022-02-10 | 2022-05-06 | 北京理工大学 | Unmanned vehicle local dynamic obstacle avoidance path planning method |
| CN114848315A (en) * | 2022-05-05 | 2022-08-05 | 广东工业大学 | Intelligent wheelchair man-machine cooperative control system based on surface electromyogram signals |
Non-Patent Citations (1)
| Title |
|---|
| 缪露: "基于双目视觉的轮椅自动跟随方法研究", pages 27 - 29 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115469665B (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Compare contact model-based control and contact model-free learning: A survey of robotic peg-in-hole assembly strategies | |
| Ren et al. | Learning inverse kinematics and dynamics of a robotic manipulator using generative adversarial networks | |
| Kim et al. | Estimating the non-linear dynamics of free-flying objects | |
| Siradjuddin et al. | Image-based visual servoing of a 7-DOF robot manipulator using an adaptive distributed fuzzy PD controller | |
| Maravall et al. | Vision-based anticipatory controller for the autonomous navigation of an UAV using artificial neural networks | |
| Gao et al. | Contextual task-aware shared autonomy for assistive mobile robot teleoperation | |
| CN115351780A (en) | Method for controlling a robotic device | |
| Chamorro et al. | Neural network based lidar gesture recognition for realtime robot teleoperation | |
| Pajak et al. | Point-to-point collision-free trajectory planning for mobile manipulators | |
| Jangid et al. | Kinematics-based end-effector path control of a mobile manipulator system on an uneven terrain using a two-stage Support Vector Machine | |
| CN113547501B (en) | SLAM-based mobile mechanical arm cart task planning and control method | |
| Patle et al. | Optimal trajectory planning of the industrial robot using hybrid S-curve-PSO approach | |
| Arai et al. | Fast prediction of a worker’s reaching motion without a skeleton model (f-premo) | |
| Jiang et al. | Learning decentralized control policies for multi-robot formation | |
| Tsai et al. | Visually guided picking control of an omnidirectional mobile manipulator based on end-to-end multi-task imitation learning | |
| Deng et al. | A learning framework for semantic reach-to-grasp tasks integrating machine learning and optimization | |
| Chaudhary et al. | Controlling a swarm of unmanned aerial vehicles using full-body k-nearest neighbor based action classifier | |
| Zhou et al. | RETRACTED: uncalibrated dynamic visual servoing via multivariate adaptive regression splines and improved incremental extreme learning machine | |
| Bócsi et al. | Learning tracking control with forward models | |
| Chow et al. | Learning human navigational skill for smart wheelchair in a static cluttered route | |
| CN115469665B (en) | Intelligent wheelchair target tracking control method and system suitable for dynamic environment | |
| Li et al. | A novel semi-autonomous teleoperation method for the tiangong-2 manipulator system | |
| Nasrallah et al. | Elevating Mobile Robotics: Pioneering Applications of Artificial Intelligence and Machine Learning. | |
| Park et al. | Digital Twin and Deep Reinforcement Learning-Driven Robotic Automation System for Confined Workspaces: A Nozzle Dam Replacement Case Study in Nuclear Power Plants | |
| Ouarda | Neural path planning for mobile robots |
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 |