CN113190018B - Intelligent agent path control method based on improved course error rate - Google Patents
Intelligent agent path control method based on improved course error rate Download PDFInfo
- Publication number
- CN113190018B CN113190018B CN202110563095.8A CN202110563095A CN113190018B CN 113190018 B CN113190018 B CN 113190018B CN 202110563095 A CN202110563095 A CN 202110563095A CN 113190018 B CN113190018 B CN 113190018B
- Authority
- CN
- China
- Prior art keywords
- intelligent agent
- path
- agent
- improved
- node
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 239000003795 chemical substances by application Substances 0.000 claims description 74
- 239000003016 pheromone Substances 0.000 claims description 18
- 230000000694 effects Effects 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 5
- 241000257303 Hymenoptera Species 0.000 claims description 3
- 230000009191 jumping Effects 0.000 claims description 3
- 230000002035 prolonged effect Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 230000006872 improvement Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 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 or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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 or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
Abstract
The invention discloses an intelligent agent path control method based on an improved course error rate, which comprises the following design steps: 1. establishing an Ackerman corner model of the intelligent agent; 2. introducing a course error rate design agent to an improved pure tracking path tracking controller; 3. the method comprises the steps of rapidly solving the optimal solution of an objective function by using an ant colony algorithm, and taking the obtained optimal solution as a parameter of a control quantity at the current moment; 4. and obtaining the current moment control quantity of the intelligent body, namely the expected tire steering angle of the intelligent body according to the obtained control quantity parameter, and controlling the intelligent body to run. According to the method, the influence of the forward-looking distance in pure tracking control is reduced by optimizing the pure tracking algorithm, and the real-time performance of intelligent agent path tracking is improved.
Description
Technical Field
The invention belongs to the technical field of robots, and particularly relates to an intelligent agent path control method based on an improved course error rate.
Background
In the field of autonomous driving, a context awareness module, a path planning module and a path tracking module are generally included. Aiming at the path tracking technology, scientific researchers deeply explore intelligent control, mainly comprising PID control, pure tracking control, Steiner control, fuzzy control, model prediction control, LQR control and the like, wherein two or more methods are even combined to optimize parameters, enhance the robustness of a control system and the like. Modeless PID controllers have advantages because the formulas are relatively simple. However, the PID controller is sensitive to errors, and the system stability is poor. Therefore, researchers research the model of the intelligent agent, such as Chen S-P and the like, and propose a high-speed automatic driving intelligent agent path tracking control method based on the combination of model prediction control and PID speed control, so that the intelligent agent can obtain the most appropriate steering angle in real time. Shaobing Xu et al designs a predictive steering control that can calculate the steering angle of the agent and control the speed required for the reference path in real time. Nitin R et al propose a feedback-feed-forward steering controller that can maintain the stability of the agent while maintaining the steering limits, while minimizing lateral trajectory tracking bias. Erkan Kayacan uses a model predictive controller to control autonomous driving of an agent. Heng yang Wang et al propose an improved model predictive control controller based on fuzzy adaptive weighting control, which not only ensures the tracking accuracy, but also considers the dynamic stability of the agent in the tracking process. The algorithm can lead the balance trolley to well track the designated track on the premise of keeping dynamic balance, has higher dynamic response speed and has good robustness to interference. However, the method relates to intelligent body dynamic modeling, the system has nonlinearity, the calculation is complex, the real-time performance is poor, and if the initial value is not appropriate, the optimization is poor, so that the method is not widely applied in practice.
Different from intelligent agent modeling, the controller designed based on the intelligent agent geometric model can directly select the track of a planned path for control, so that the controller has a simple structure and is convenient to control and widely applied in practice. Pure tracking control is the most applied geometric model. Qiao N et al propose a pure tracking algorithm based on neutral angle adaptive calibration, and optimize its parameters with a particle swarm algorithm. However, pure tracking controllers require the selection of an appropriate look-ahead distance, so Chen, I-Ming et al, at Berkeley university, California, propose a recent policy optimization algorithm in combination with a pure tracking method to construct an intelligent controller, and AbdElmoniem et al propose a Stanley improvement algorithm based on predicting future states of the intelligent. Lingli Yu et al propose an improved pure tracking algorithm based on fuzzy control, which improves the stability of the pure tracking algorithm. Yunxiao shann et al propose to improve pure tracking control by replacing the circle with a clothoid curve, which can improve the curve fit. However, although these methods solve the drawback of forward looking distance, the structure of the controller is greatly modified, and the controller implementation is complicated.
Although these studies have gained some use, there is still a challenge with pure tracking control, namely the uncertainty of the effect of look-ahead distance on pure tracking control. Particularly for automatic control of an intelligent agent, the path tracking performance of the intelligent agent mainly depends on real-time control of steering, and the traditional pure tracking path tracking method has poor dynamic property in actual engineering, so that the tracking precision is influenced. Therefore, there is a need to improve the real-time performance of path tracing by improving pure tracing algorithms.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the problems of insufficient forward looking distance, low initial convergence speed and the like of the traditional pure tracking control, the invention provides an intelligent agent path control method based on improved course error rate, which comprises the following specific steps:
step 1, establishing an Ackerman corner model of an agent:
where δ is the desired front wheel steering angle, l represents the agent wheelbase, and R is the radius of the circular trajectory that the agent follows for a given front wheel steering angle δ;
step 2, designing an improved pure tracking controller of the intelligent agent, and improving the algorithm by introducing a course error rate, wherein the method comprises the following steps:
(2.1) establishing a pure tracking control model of the intelligent agent:
wherein: delta is the desired front wheel steering angle, L represents the smart body wheelbase, LdRepresenting look-ahead distance, α representing the current direction and the difference in direction to the look-ahead point;
(2.2) calculating a course error theta of the intelligent agent according to the current state of the intelligent agent;
(2.3) establishing an improved pure tracking control model:
wherein k is1、k2Is a parameter;
step 3, quickly solving k in step 2.3 by utilizing ant colony algorithm1And k2Predicting the optimal state of the agent at the next moment, and calculating the optimal parameter k at the current moment1opt、k2opt;
Step 4, k of the current time is compared1opt、k2optSubstituting the optimal gain parameter into the improved pure tracking control model of the intelligent agent to obtain the optimal expected steering angle delta of the intelligent agentoptControlling the operation of the intelligent agent; and judging whether the reference point is required to be switched or not, and jumping to the step 2.3 to continue the control of the next moment until the intelligent agent reaches the end point of the reference path.
As a further improvement of the invention, the intelligent agent is an unmanned robot or an unmanned vehicle.
As a further improvement of the present invention, the step 3 comprises:
(3.1) according to k1And k2Selecting digit number according to the value range and precision, wherein the integer digit number is 1 and the decimal digit number is 4 according to experience; the moving path of each ant is more than or equal to 1 and less than or equal to 10, and y is more than or equal to 0 and less than or equal to 9;
(3.2) initializing ant colony parameters: setting ant colony scale, pheromone volatilization coefficient, pheromone intensity, maximum iteration times, importance degree of remaining pheromone and importance degree of elicitation information;
(3.3) at tjAt that moment, the probability that the kth ant selects the next node is
Wherein: τ (i, y)i,tj) Denotes the concentration of the pheromone at node I, η (I, y)i,tj) Indicating the visibility of the information on the node I,whereinRepresenting the position of each node on the optimal path in the current state;
(3.4) giving the transition probability PcIf, ifThen it means that node l canTo serve as the next transfer node;
(3.5) calculating the objective function value of the ant after all ants finish one-time path optimization;
(3.6) updating the pheromone concentration of the node, wherein rho is a volatility coefficient, the value of rho is related to the convergence rate and the global search capacity of the ant colony algorithm, and if the value is too small, the larger the positive feedback effect of information is, the smaller the search randomness is, and the easier the search is trapped in local convergence; on the contrary, if the value of rho is too large, the search randomness is enhanced, the convergence time is prolonged, and the search efficiency is influenced.The pheromone increment represented as a node I is calculated as
Wherein L iskRepresents the path length of the kth ant;
(3.7) entering next circulation, continuously repeating the steps, obtaining an optimal path when the iteration times reach a preset maximum value, and obtaining the optimal path according to the formula
To obtain k1、k2Is optimized parameter k1optAnd k2opt。
As a further improvement of the invention, the step (4) comprises the following steps: obtaining the optimal parameter k by ant colony algorithm1optAnd k2optAnd (3) introducing an improved pure tracking algorithm, and calculating the current moment control quantity of the intelligent agent:
the optimal steering angle delta of the intelligent agent at the current moment can be obtainedoptTo control the operation of the agent.
Has the advantages that: compared with the prior art, the intelligent agent path self-adaptive control method and the intelligent agent path self-adaptive control system based on the improved course error rate have the following advantages: the algorithm parameters are optimized and adjusted by utilizing the ant colony, the searching capability of the system is enhanced, the defects of the traditional pure tracking control can be overcome, the controller is simple in structure and easy to realize, and the method can be effectively applied to the intelligent self-adaptive path tracking control.
Drawings
FIG. 1 is a flow chart of a method for intelligent agent path control based on improved course error rate in accordance with the present disclosure;
fig. 2 is a flowchart for calculating the optimal parameters at the current time by using the ant colony algorithm.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention are described below with reference to the accompanying drawings.
The invention discloses an intelligent agent path control method based on an improved course error rate, which can adaptively adjust and improve a pure tracking control model according to the operating state of an intelligent agent, obtain an optimal intelligent agent steering angle, ensure the intelligent agent path tracking effect and realize the accurate operation of the intelligent agent.
As an embodiment of the present invention, the present invention discloses an intelligent agent path control method based on improved course error rate, wherein the intelligent agent is an unmanned robot or an unmanned vehicle, and a flow chart is shown in fig. 1, and the method comprises the following steps:
step 1, establishing an ackermann corner model of an agent:
where δ is the desired front wheel steering angle, l represents the agent wheelbase, and R is the radius of the circular trajectory that the agent follows for a given front wheel steering angle δ;
step 2, designing an improved pure tracking controller of the intelligent agent, and improving the algorithm by introducing a course error rate, wherein the method comprises the following steps:
(2.1) establishing a pure tracking control model of the intelligent agent:
wherein: delta is the desired front wheel steering angle, L represents the intelligent body wheelbase, LdRepresenting the look-ahead distance, α representing the current direction and the difference in direction to the look-ahead point;
(2.2) calculating a course error theta of the intelligent agent according to the current state of the intelligent agent;
(2.3) establishing an improved pure tracking control model:
wherein k is1、k2Is a parameter;
step 3, quickly solving k in step 2.3 by utilizing ant colony algorithm1And k2The flow chart of the ant colony algorithm for calculating the optimal parameter at the current moment is shown in fig. 2, the optimal state of the intelligent agent at the next moment is predicted, and the optimal parameter k at the current moment is calculated1opt、k2opt;
The step 3 comprises the following steps:
(3.1) according to k1And k2Selecting digit number according to the value range and precision, wherein the integer digit number is 1 and the decimal digit number is 4 according to experience; the moving path of each ant is more than or equal to 1 and less than or equal to 10, and y is more than or equal to 0 and less than or equal to 9;
(3.2) initializing ant colony parameters: setting ant colony scale, pheromone volatilization coefficient, pheromone intensity, maximum iteration times, importance degree of remaining pheromone and importance degree of elicitation information;
(3.3) at tjAt that moment, the probability that the kth ant selects the next node is
Wherein: τ (i, y)i,tj) Indicates the pheromone concentration on node I, η (I, y)i,tj) Indicating the visibility of the information on the node I,whereinRepresenting the position of each node on the optimal path in the current state;
(3.5) calculating the objective function value of the ant after all ants finish one-time path optimization;
(3.6) updating the pheromone concentration of the node, wherein rho is a volatility coefficient, the value of rho is related to the convergence rate and the global search capacity of the ant colony algorithm, and if the value is too small, the larger the positive feedback effect of information is, the smaller the search randomness is, and the easier the search is trapped in local convergence; on the contrary, if the value of rho is too large, the search randomness is enhanced, the convergence time is prolonged, and the search efficiency is influenced.The pheromone increment represented as a node I is calculated as
Wherein L iskRepresents the path length of the kth ant;
(3.7) entering next circulation, continuously repeating the steps, obtaining an optimal path when the iteration times reach a preset maximum value, and obtaining the optimal path according to the formula
To obtain k1、k2Is optimized parameter k1optAnd k2opt。
Step 4, k of the current time is compared1opt、k2optSubstituting the optimal gain parameter into the improved pure tracking control model of the intelligent agent to obtain the optimal expected steering angle delta of the intelligent agentoptControlling the operation of the intelligent agent; and judging whether the reference point is required to be switched or not, and jumping to the step 2.3 to continue the control of the next moment until the intelligent agent reaches the end point of the reference path.
The step (4) comprises the following steps: obtaining the optimal parameter k by the ant colony algorithm1optAnd k2optAnd (3) introducing an improved pure tracking algorithm, and calculating the current moment control quantity of the intelligent agent:
the optimal steering angle delta of the intelligent agent at the current moment can be obtainedoptTo control the operation of the agent.
The above description is only one of the preferred embodiments of the present invention, and is not intended to limit the present invention in any way, but any modifications or equivalent variations made in accordance with the technical spirit of the present invention are within the scope of the present invention as claimed.
Claims (3)
1. An intelligent agent path control method based on improved course error rate comprises the following specific steps:
step 1, establishing an ackermann corner model of an agent:
where δ is the desired front wheel steering angle, l represents the agent wheelbase, and R is the radius of the circular trajectory that the agent follows for a given front wheel steering angle δ;
step 2, designing an improved pure tracking controller of the intelligent agent, and improving an algorithm by introducing a course error rate, wherein the method comprises the following steps:
(2.1) establishing a pure tracking control model of the intelligent agent:
wherein: delta is the desired front wheel steering angle, L represents the smart body wheelbase, LdRepresenting look-ahead distance, α representing the current direction and the difference in direction to the look-ahead point;
(2.2) calculating a course error theta of the intelligent agent according to the current state of the intelligent agent;
(2.3) establishing an improved pure tracking control model:
wherein k is1、k2Is a parameter;
step 3, quickly solving k in step 2.3 by utilizing ant colony algorithm1And k2Predicting the optimal state of the agent at the next moment, and calculating the optimal parameter k at the current moment1opt、k2opt;
The step 3 comprises the following steps:
(3.1) according to k1And k2Selecting digit according to the value range and precision, wherein the integer digit is 1 and the decimal digit is 4 according to experience; the moving path of each ant is more than or equal to 1 and less than or equal to 10, and y is more than or equal to 0 and less than or equal to 9;
(3.2) initializing ant colony parameters: setting ant colony scale, pheromone volatilization coefficient, pheromone intensity, maximum iteration times, the importance degree of remaining pheromone and the importance degree of inspiring information;
(3.3) at tjAt that moment, the probability that the kth ant selects the next node is
Wherein: τ (i, y)i,tj) Indicates the pheromone concentration on node I, η (I, y)i,tj) Indicating the visibility of the information on the node I,whereinRepresenting the position of each node on the optimal path in the current state;
(3.5) calculating the objective function value of the ant after all ants finish one-time path optimization;
(3.6) updating the pheromone concentration of the node, wherein rho is a volatility coefficient, the value of rho is related to the convergence rate and the global search capacity of the ant colony algorithm, and if the value is too small, the larger the positive feedback effect of information is, the smaller the search randomness is, and the easier the search is trapped in local convergence; on the contrary, if the value of rho is too large, the search randomness is enhanced, the convergence time is prolonged, the search efficiency is influenced,the pheromone increment represented as a node I is calculated as
Wherein L iskRepresents the path length of the kth ant;
(3.7) entering next circulation, continuously repeating the steps, obtaining an optimal path when the iteration times reach a preset maximum value, and obtaining the optimal path according to the formula
To obtain k1、k2Is optimized parameter k1optAnd k2opt;
Step 4, k of the current time is compared1opt、k2optSubstituting the optimal gain parameter into the improved pure tracking control model of the intelligent agent to obtain the optimal expected steering angle delta of the intelligent agentoptControlling the operation of the intelligent agent; and judging whether the reference point is required to be switched or not, and jumping to the step 2.3 to continue the control of the next moment until the intelligent agent reaches the end point of the reference path.
2. An intelligent agent path control method based on improved course error rate according to claim 1, wherein the intelligent agent is an unmanned robot or an unmanned vehicle.
3. An intelligent agent path control method based on improved heading error rate according to claim 1, wherein said step 4 comprises: obtaining the optimal parameter k by the ant colony algorithm1optAnd k2optAnd (3) introducing an improved pure tracking algorithm, and calculating the current moment control quantity of the intelligent agent:
the current moment of the intelligent agent can be obtainedOptimum steering angle deltaoptTo control the operation of the agent.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110563095.8A CN113190018B (en) | 2021-05-24 | 2021-05-24 | Intelligent agent path control method based on improved course error rate |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110563095.8A CN113190018B (en) | 2021-05-24 | 2021-05-24 | Intelligent agent path control method based on improved course error rate |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113190018A CN113190018A (en) | 2021-07-30 |
CN113190018B true CN113190018B (en) | 2022-07-01 |
Family
ID=76984816
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110563095.8A Active CN113190018B (en) | 2021-05-24 | 2021-05-24 | Intelligent agent path control method based on improved course error rate |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113190018B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113978550B (en) * | 2021-12-30 | 2022-03-25 | 北京慧拓无限科技有限公司 | Reversing control method for unmanned mine car |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101501699A (en) * | 2005-04-08 | 2009-08-05 | 里卡多公司 | Vehicle chassis and powertrain set up tool for track trajectory and speed optimization |
CN105867377B (en) * | 2016-04-13 | 2019-01-08 | 浙江理工大学 | A kind of automatic steering control of farm mechanism control method |
CN108791491A (en) * | 2018-06-12 | 2018-11-13 | 中国人民解放军国防科技大学 | Vehicle lateral tracking control method based on self-evaluation learning |
CN110362084B (en) * | 2019-07-22 | 2022-08-26 | 东南大学 | Vehicle path tracking control method based on ant colony algorithm |
CN111176284B (en) * | 2020-01-02 | 2022-09-23 | 东南大学 | Self-adaptive control method and system for vehicle path tracking for unmanned driving |
GB2594456A (en) * | 2020-04-27 | 2021-11-03 | Caterpillar Inc | System and method for autonomous steering calibration |
-
2021
- 2021-05-24 CN CN202110563095.8A patent/CN113190018B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN113190018A (en) | 2021-07-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110362084B (en) | Vehicle path tracking control method based on ant colony algorithm | |
CN111273551A (en) | Double-steering wheel automatic guided vehicle trajectory tracking control method and system | |
CN111176284B (en) | Self-adaptive control method and system for vehicle path tracking for unmanned driving | |
CN110989577B (en) | Automatic driving decision method and automatic driving device of vehicle | |
CN109884900A (en) | The design method of cropper path following control device based on adaptive model PREDICTIVE CONTROL | |
CN109933057B (en) | Local guide track planning method and device for automatic tractor driving system | |
CN111552183B (en) | Six-legged robot obstacle avoidance method based on adaptive weight reinforcement learning | |
CN113190018B (en) | Intelligent agent path control method based on improved course error rate | |
CN111538328A (en) | Priority hierarchical prediction control method for obstacle avoidance trajectory planning and tracking control of autonomous driving vehicle | |
CN112222703B (en) | Energy consumption optimal trajectory planning method for welding robot | |
CN110687797B (en) | Self-adaptive MPC parking transverse control method based on position and posture | |
CN110789530B (en) | Four-wheel independent steering-independent driving vehicle trajectory tracking method and system | |
CN110471289A (en) | A kind of the Adaptive Path tracking and system of view-based access control model navigation mobile device | |
Zhu et al. | An adaptive path tracking method for autonomous land vehicle based on neural dynamic programming | |
CN116661469B (en) | Robot track error correction method and system | |
CN115562265A (en) | Mobile robot path planning method based on improved A-x algorithm | |
Wang et al. | An improved pure pursuit path tracking control method based on heading error rate | |
CN113703450A (en) | Mobile robot path planning method for improving ant colony algorithm based on smooth factors | |
CN111221338B (en) | Path tracking method, device, equipment and storage medium | |
CN112099515A (en) | Automatic driving method for lane change avoidance | |
Xie et al. | Parameter self‐learning feedforward compensation‐based active disturbance rejection for path‐following control of self‐driving forklift trucks | |
Xiao et al. | Multiobjective path optimization of an indoor AGV based on an improved ACO-DWA | |
CN109141438B (en) | Forklift global path planning method | |
Wu et al. | Steering wheel AGV path tracking control based on improved pure pursuit model | |
Zhang et al. | Using partial-policy q-learning to plan path for robot navigation in unknown enviroment |
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 |