CN108984275A - The agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study - Google Patents
The agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study Download PDFInfo
- Publication number
- CN108984275A CN108984275A CN201810980718.XA CN201810980718A CN108984275A CN 108984275 A CN108984275 A CN 108984275A CN 201810980718 A CN201810980718 A CN 201810980718A CN 108984275 A CN108984275 A CN 108984275A
- Authority
- CN
- China
- Prior art keywords
- scene
- unity3d
- farmland
- agricultural
- training
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45504—Abstract machines for programme code execution, e.g. Java virtual machine [JVM], interpreters, emulators
- G06F9/45508—Runtime interpretation or emulation, e g. emulator loops, bytecode interpretation
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
It include a variety of environment parameters in classical farmland scene based on the agricultural driver training method of Intelligent unattended that Unity3D and depth enhancing learn, including step 1, the classical farmland scene of foundation in Unity3D;Step 2 is based on classical farmland scene progress simulation operations by driver, and by operating process parametrization, obtains operation parameter;Step 3 constructs convolutional neural networks according to environment parameter and operation parameter;Step 4 carries out pre-training to convolutional neural networks, obtains pre-training intelligent body;Step 5, random generation dynamic farmland scene, and random environment parameter is added in the scene of dynamic farmland;Step 6 puts into pre-training intelligent body in the scene of dynamic farmland, carries out self training using depth enhancing learning algorithm, strengthened intelligent body;Step 7, in the scene of dynamic farmland, the frequency of lift portion random environment parameter, to strengthen intelligent body carry out repetition training.The present invention can fast and efficiently improve the driving ability of driver.
Description
Technical field
The present invention relates to the unmanned technical fields of agricultural machinery, are specifically learnt based on Unity3D and depth enhancing
The agricultural driver training method of Intelligent unattended.
Background technique
With the development of science and technology, agricultural machinery is increasingly intended to intelligence, automatic driving of agricultural machinery technology is efficient
The key technology of agricultural.Natural environment locating for actual production life agricultural machinery has many unknown problems to need to handle, and agricultural machinery exists
It can be potentially encountered the barriers such as electric pole, people and other machinery when operation or close to the edge of a field, agricultural machinery needs made automatically at this time
Avoidance decision.Different farmlands needs different path plannings again, to enable the more efficient operation of agricultural machinery.Thus could
Accomplish the safeguard protection to people and agricultural machinery, while again can the maximum production efficiency for playing independent navigation agricultural vehicle.How to allow
Agricultural machinery adapts to various unexpected complex scenes, is that intelligent agricultural machinery securely achieves a key of autonomous unmanned operation and asks
Topic.
In recent years, deeply study is swift and violent in academia's development, shows in the control task especially in complex environment
It is original.For deep learning in many traditional identification missions, discrimination all obtains significant raising.Also it tastes in many other fields
Deep learning on probation solves the problems, such as some this fields.Deep learning, which is applied, has had some grind in the application of object detection
Study carefully, especially and the combination of intensified learning, presents its unique advantage.Deeply study is deep learning and intensified learning
The field combined, it can be realized the method completely new from the one kind for the end-to-end study for perceiving movement.Briefly,
It is exactly to input perception information such as vision as the mankind, then passes through deep neural network, direct output action.
(the unpiloted obstacle-avoiding route planning of agricultural machinery and its controlling party to be used in Chinese patent CN201710156019.9
Method) in disclose a kind of unpiloted obstacle-avoiding route planning of agricultural machinery and its control method, pass through sensor and obtain agricultural machinery environment
Information makes avoidance decision, calculates a theoretical avoidance path using improved most chopped collimation method, excellent using the path of curve
Change method optimum theory avoidance path obtains practical avoidance path.This barrier-avoiding method can only be accomplished to detect whether that there are barriers
And simple detour avoidance is carried out, and can only accomplish to be evaded for a certain static-obstacle thing in part.This avoidance side
The mode of method obstacle avoidance is excessively single, and degree of intelligence is shallower.
Summary of the invention
In order to solve deficiency in the prior art, the present invention provides a kind of intelligence based on Unity3D and depth enhancing study
Can nobody agricultural driver training method, intelligent body can be made to carry out agricultural machinery driver training in computer environment, and provide greatly
The complex environment of amount improves the driving ability of intelligent body.
To achieve the goals above, the present invention use the specific scheme is that based on Unity3D and depth enhancing study intelligence
Can nobody agricultural driver training method, include the following steps:
Step 1 establishes classical farmland scene in Unity3D, includes a variety of environment parameters in classical farmland scene;
Step 2 is based on classical farmland scene progress simulation operations by driver, and by operating process parametrization, obtains operation ginseng
Amount;
Step 3 constructs convolutional neural networks according to environment parameter and operation parameter;
Step 4 carries out pre-training to convolutional neural networks, obtains pre-training intelligent body;
Step 5, random generation dynamic farmland scene, and random environment parameter is added in the scene of dynamic farmland;
Step 6 puts into pre-training intelligent body in the scene of dynamic farmland, carries out self training using depth enhancing learning algorithm,
Strengthened intelligent body;
Step 7, in the scene of dynamic farmland, the frequency of lift portion random environment parameter, to strengthen intelligent body carry out repeat instruction
Practice.
In the step 1, environment parameter includes weather, landforms, scene and working specification.
In the step 2, operation parameter includes throttle movement, brake and steering wheel movement.
In the step 3, using environment parameter as the input of convolutional neural networks, and be denoted as X, using operate parameter as
The output of convolutional neural networks, and it is denoted as y, driver operational data is denoted as t, then the square error of convolutional neural networks is
In the step 4, the method for carrying out pre-training to convolutional neural networks includes:
Step 4.1, the partial derivative for calculating convolutional neural networks weight w according to chain rule using square error E, specific method areWherein o is the output of neuron, and net is the input of neuron;
Step 4.2 introduces learning rate α, is updated according to the partial derivative of square error and weight to weight w, specific method
For
In the step 5, random environment parameter further includes moving obstacle, Changes in weather and traffic limitation.
In the step 6, include: using the method that depth enhancing learning algorithm carries out self training
Step 6.1, construction strategy network, using random environment parameter as the input of tactful network, with intelligent body in order to cope with
The operation that machine environment parameter is made is as output;
Reward mechanism is added in tactful network in step 6.2;
Step 6.3, according to the convolutional neural networks structure setting Actor network and Critic network in pre-training model;
Random parameter is added by Ornstein-Uhlenbeck process in step 6.4;
Step 6.5 is updated Actor network by experience replay.
In the step 6.1, tactful network is expressed as πθ(s, α)=P [α | s, θ].
In the step 6.2, attenuation coefficient γ is introduced, r is the reward value of each state, and total reward mechanism is expressed as R
=r1+γ2r2+......γnrn。
In the step 6.4, dx is expressed as by the method that random parameter is added in Ornstein-Uhlenbeck processt=
θ(μ-xt)dt+σdWt, θ expression speed of the variable to Change in Mean therein, μ expression mean value, σ is the freedom degree of process, x and W
Represent the control amount of process effect.
The utility model has the advantages that the present invention carries out the more scene farmland emulation of polymorphic type by Unity3D platform, and strong based on depth
Changing learning algorithm enables intelligent body to obtain the experiences of a large amount of scenes in a computer, improves intelligent body working efficiency and increases and is
The stability of system, to prevent from encountering the case where emergency situations can not be handled in reality scene.
Detailed description of the invention
Fig. 1 is flow chart of the invention;
Fig. 2 is convolutional neural networks structural schematic diagram of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Fig. 1 and 2 is please referred to, Fig. 1 is flow chart of the invention, and Fig. 2 is convolutional neural networks structural schematic diagram of the invention.
The agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study, including step 1 is to 7.
Step 1 establishes classical farmland scene in Unity3D, includes a variety of environment parameters, environment in classical farmland scene
Parameter includes weather, landforms, scene and working specification.
Step 2 is based on classical farmland scene progress simulation operations by driver, and by operating process parametrization, is grasped
Make parameter, operation parameter includes throttle movement, brake and steering wheel movement.
Step 3, according to environment parameter and operation parameter construct convolutional neural networks, specifically, using environment parameter as
The input of convolutional neural networks, and it is denoted as X, to operate the output of parameter as convolutional neural networks, and it is denoted as y, will driven
The person's of sailing operation data is denoted as t, then the square error of convolutional neural networks is
Step 4 carries out pre-training to convolutional neural networks, obtains pre-training intelligent body, specific method include step 4.1 to
4.2。
Step 4.1, the partial derivative for calculating convolutional neural networks weight w according to chain rule using square error E, specific side
Method isWherein o is the output of neuron, and net is the input of neuron.
Step 4.2 introduces learning rate α, is updated according to the partial derivative of square error and weight to weight w, specifically
Method is
Step 5, random generation dynamic farmland scene, and random environment parameter is added in the scene of dynamic farmland, at random
Environment parameter further includes moving obstacle, Changes in weather and traffic limitation.
Step 6 puts into pre-training intelligent body in the scene of dynamic farmland, carries out self instruction using depth enhancing learning algorithm
Practice, strengthened intelligent body, and specific method includes step 6.1 to 6.5.
It include step 6.1 to 6.5 using the method that depth enhancing learning algorithm carries out self training in step 6.
Step 6.1, construction strategy network, using random environment parameter as the input of tactful network, with intelligent body in order to answer
Output is used as to the operation that random environment parameter is made, tactful network is expressed as πθ(s, α)=P [α | s, θ].
Reward mechanism is added in step 6.2 in tactful network, introduces attenuation coefficient γ, and r is the reward value of each state,
Total reward mechanism is expressed as R=r1+γ2r2+......γnrn。
Step 6.3, according to the convolutional neural networks structure setting Actor network and Critic network in pre-training model.
Random parameter is added by Ornstein-Uhlenbeck process in step 6.4, and specific method is expressed as dxt=θ
(μ-xt)dt+σdWt, θ expression speed of the variable to Change in Mean therein, μ expression mean value, σ is the freedom degree of process, x and W generation
The control amount of table process effect.
Step 6.5 is updated Actor network by experience replay.
Step 7, in the scene of dynamic farmland, the frequency of lift portion random environment parameter, to strengthen intelligent body carry out weight
Refreshment is practiced.It is specifically directed in reality and is not easy to acquire data, but incidental dangerous scene carries out a large amount of repetition trainings,
So that intelligent body grasps the ability for coping with such scene.
The present invention carries out the more scene farmland emulation of polymorphic type by Unity3D platform, and is calculated based on deeply study
Method enables intelligent body to obtain the experience of a large amount of scenes in a computer, improves intelligent body working efficiency and increases the stabilization of system
Property, to prevent from encountering the case where emergency situations can not be handled in reality scene.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (10)
1. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study, it is characterised in that: including as follows
Step:
Step 1 establishes classical farmland scene in Unity3D, includes a variety of environment parameters in classical farmland scene;
Step 2 is based on classical farmland scene progress simulation operations by driver, and by operating process parametrization, obtains operation ginseng
Amount;
Step 3 constructs convolutional neural networks according to environment parameter and operation parameter;
Step 4 carries out pre-training to convolutional neural networks, obtains pre-training intelligent body;
Step 5, random generation dynamic farmland scene, and random environment parameter is added in the scene of dynamic farmland;
Step 6 puts into pre-training intelligent body in the scene of dynamic farmland, carries out self training using depth enhancing learning algorithm,
Strengthened intelligent body;
Step 7, in the scene of dynamic farmland, the frequency of lift portion random environment parameter, to strengthen intelligent body carry out repeat instruction
Practice.
2. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as described in claim 1,
Be characterized in that: in the step 1, environment parameter includes weather, landforms, scene and working specification.
3. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as described in claim 1,
Be characterized in that: in the step 2, operation parameter includes throttle movement, brake and steering wheel movement.
4. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as described in claim 1,
It is characterized in that: in the step 3, using environment parameter as the input of convolutional neural networks, and being denoted as X, to operate parameter work
For the output of convolutional neural networks, and it is denoted as y, driver operational data is denoted as t, then the square error of convolutional neural networks
For
5. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as claimed in claim 4,
Be characterized in that: in the step 4, the method for carrying out pre-training to convolutional neural networks includes:
Step 4.1, the partial derivative for calculating convolutional neural networks weight w according to chain rule using square error E, specific method are
Wherein o is the output of neuron, and net is the input of neuron;
Step 4.2 introduces learning rate α, is updated according to the partial derivative of square error and weight to weight w, specific method
For
6. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as claimed in claim 5,
Be characterized in that: in the step 5, random environment parameter further includes moving obstacle, Changes in weather and traffic limitation.
7. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as claimed in claim 6,
It is characterized in that: in the step 6, including: using the method that depth enhancing learning algorithm carries out self training
Step 6.1, construction strategy network, using random environment parameter as the input of tactful network, with intelligent body in order to cope with
The operation that machine environment parameter is made is as output;
Reward mechanism is added in tactful network in step 6.2;
Step 6.3, according to the convolutional neural networks structure setting Actor network and Critic network in pre-training model;
Random parameter is added by Ornstein-Uhlenbeck process in step 6.4;
Step 6.5 is updated Actor network by experience replay.
8. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as claimed in claim 7,
Be characterized in that: in the step 6.1, tactful network is expressed as πθ(s, α)=P [α | s, θ], wherein s is current state, and α is institute
The behavior taken, θ are policing parameter.
9. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as claimed in claim 8,
It is characterized in that: in the step 6.2, introducing attenuation coefficient γ, r is the reward value of each state, and total reward mechanism is expressed as R
=r1+γ2r2+......γnrn。
10. the agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study as claimed in claim 9,
It is characterized by: being expressed as in the step 6.4 by the method that random parameter is added in Ornstein-Uhlenbeck process
dxt=θ (μ-xt)dt+σdWt, θ expression speed of the variable to Change in Mean therein, μ expression mean value, σ is the freedom degree of process,
X and W represents the control amount of process effect.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810980718.XA CN108984275A (en) | 2018-08-27 | 2018-08-27 | The agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810980718.XA CN108984275A (en) | 2018-08-27 | 2018-08-27 | The agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108984275A true CN108984275A (en) | 2018-12-11 |
Family
ID=64547096
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810980718.XA Pending CN108984275A (en) | 2018-08-27 | 2018-08-27 | The agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108984275A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109946995A (en) * | 2019-03-26 | 2019-06-28 | 湖北亿咖通科技有限公司 | Emulation test method, device and the intelligent terminal of automatic Pilot |
CN110824954A (en) * | 2019-10-24 | 2020-02-21 | 北京仿真中心 | Intelligent agent training method and system, computer equipment and readable storage medium |
CN113706725A (en) * | 2021-08-31 | 2021-11-26 | 的卢技术有限公司 | Method for realizing mobile environment based on unity ML plug-in |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106842925A (en) * | 2017-01-20 | 2017-06-13 | 清华大学 | A kind of locomotive smart steering method and system based on deeply study |
CN106970615A (en) * | 2017-03-21 | 2017-07-21 | 西北工业大学 | A kind of real-time online paths planning method of deeply study |
CN106981232A (en) * | 2017-03-27 | 2017-07-25 | 江苏科技大学 | A kind of virtual driving simulator based on unity3D |
CN107392189A (en) * | 2017-09-05 | 2017-11-24 | 百度在线网络技术(北京)有限公司 | For the method and apparatus for the driving behavior for determining unmanned vehicle |
CN107491072A (en) * | 2017-09-05 | 2017-12-19 | 百度在线网络技术(北京)有限公司 | Vehicle obstacle-avoidance method and apparatus |
CN107564363A (en) * | 2017-09-05 | 2018-01-09 | 百度在线网络技术(北京)有限公司 | A kind of method and apparatus for driving model switching |
CN107817798A (en) * | 2017-10-30 | 2018-03-20 | 洛阳中科龙网创新科技有限公司 | A kind of farm machinery barrier-avoiding method based on deep learning system |
US20180129974A1 (en) * | 2016-11-04 | 2018-05-10 | United Technologies Corporation | Control systems using deep reinforcement learning |
US20180165602A1 (en) * | 2016-12-14 | 2018-06-14 | Microsoft Technology Licensing, Llc | Scalability of reinforcement learning by separation of concerns |
CN108255182A (en) * | 2018-01-30 | 2018-07-06 | 上海交通大学 | A kind of service robot pedestrian based on deeply study perceives barrier-avoiding method |
CN108268887A (en) * | 2017-12-15 | 2018-07-10 | 西安电子科技大学 | Driver's awareness of safety appraisal procedure based on virtual driving and EEG detections |
CN108333959A (en) * | 2018-03-09 | 2018-07-27 | 清华大学 | A kind of energy saving method of operating of locomotive based on convolutional neural networks model |
CN108427961A (en) * | 2018-02-11 | 2018-08-21 | 陕西师范大学 | Synthetic aperture focusing imaging depth appraisal procedure based on convolutional neural networks |
-
2018
- 2018-08-27 CN CN201810980718.XA patent/CN108984275A/en active Pending
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180129974A1 (en) * | 2016-11-04 | 2018-05-10 | United Technologies Corporation | Control systems using deep reinforcement learning |
US20180165602A1 (en) * | 2016-12-14 | 2018-06-14 | Microsoft Technology Licensing, Llc | Scalability of reinforcement learning by separation of concerns |
CN106842925A (en) * | 2017-01-20 | 2017-06-13 | 清华大学 | A kind of locomotive smart steering method and system based on deeply study |
CN106970615A (en) * | 2017-03-21 | 2017-07-21 | 西北工业大学 | A kind of real-time online paths planning method of deeply study |
CN106981232A (en) * | 2017-03-27 | 2017-07-25 | 江苏科技大学 | A kind of virtual driving simulator based on unity3D |
CN107392189A (en) * | 2017-09-05 | 2017-11-24 | 百度在线网络技术(北京)有限公司 | For the method and apparatus for the driving behavior for determining unmanned vehicle |
CN107491072A (en) * | 2017-09-05 | 2017-12-19 | 百度在线网络技术(北京)有限公司 | Vehicle obstacle-avoidance method and apparatus |
CN107564363A (en) * | 2017-09-05 | 2018-01-09 | 百度在线网络技术(北京)有限公司 | A kind of method and apparatus for driving model switching |
CN107817798A (en) * | 2017-10-30 | 2018-03-20 | 洛阳中科龙网创新科技有限公司 | A kind of farm machinery barrier-avoiding method based on deep learning system |
CN108268887A (en) * | 2017-12-15 | 2018-07-10 | 西安电子科技大学 | Driver's awareness of safety appraisal procedure based on virtual driving and EEG detections |
CN108255182A (en) * | 2018-01-30 | 2018-07-06 | 上海交通大学 | A kind of service robot pedestrian based on deeply study perceives barrier-avoiding method |
CN108427961A (en) * | 2018-02-11 | 2018-08-21 | 陕西师范大学 | Synthetic aperture focusing imaging depth appraisal procedure based on convolutional neural networks |
CN108333959A (en) * | 2018-03-09 | 2018-07-27 | 清华大学 | A kind of energy saving method of operating of locomotive based on convolutional neural networks model |
Non-Patent Citations (2)
Title |
---|
刘全 等: "深度强化学习综述", 《计算机学报》 * |
翟建伟: "基于深度Q网络算法与模型的研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109946995A (en) * | 2019-03-26 | 2019-06-28 | 湖北亿咖通科技有限公司 | Emulation test method, device and the intelligent terminal of automatic Pilot |
CN110824954A (en) * | 2019-10-24 | 2020-02-21 | 北京仿真中心 | Intelligent agent training method and system, computer equipment and readable storage medium |
CN113706725A (en) * | 2021-08-31 | 2021-11-26 | 的卢技术有限公司 | Method for realizing mobile environment based on unity ML plug-in |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yang et al. | A new robot navigation algorithm based on a double-layer ant algorithm and trajectory optimization | |
CN104298239B (en) | A kind of indoor mobile robot strengthens map study paths planning method | |
CN108984275A (en) | The agricultural driver training method of Intelligent unattended based on Unity3D and depth enhancing study | |
CN105137967B (en) | The method for planning path for mobile robot that a kind of depth autocoder is combined with Q learning algorithms | |
CN106022463B (en) | Based on the individualized learning method for optimizing route for improving particle cluster algorithm | |
CN106843235A (en) | It is a kind of towards the Artificial Potential Field path planning without person bicycle | |
CN106444738A (en) | Mobile robot path planning method based on dynamic motion primitive learning model | |
CN104317293B (en) | City rescue intelligent agent dynamic path planning method based on improved ant colony algorithm | |
CN108444489A (en) | A kind of paths planning method improving RRT algorithms | |
CN106441308A (en) | Unmanned aerial vehicle path planning method based on adaptive weight pigeon flock algorithm | |
CN108037758A (en) | A kind of method for planning path for mobile robot based on improvement AFSA | |
CN106705970A (en) | Multi-UAV(Unmanned Aerial Vehicle) cooperation path planning method based on ant colony algorithm | |
CN110488872A (en) | A kind of unmanned plane real-time route planing method based on deeply study | |
CN108180914A (en) | A kind of method for planning path for mobile robot improved based on ant colony with despiking | |
CN110472738A (en) | A kind of unmanned boat Real Time Obstacle Avoiding algorithm based on deeply study | |
CN106022471A (en) | Wavelet neural network model ship rolling real-time prediction method based on particle swarm optimization algorithm | |
CN113268081B (en) | Small unmanned aerial vehicle prevention and control command decision method and system based on reinforcement learning | |
CN108594804A (en) | Automatic driving control method for distribution trolley based on deep Q network | |
CN107562053A (en) | A kind of Hexapod Robot barrier-avoiding method based on fuzzy Q-learning | |
CN104503453A (en) | Mobile robot path planning method based on bacterial foraging potential field method | |
CN109947129A (en) | Rotor wing unmanned aerial vehicle paths planning method based on Dijkstra and improvement particle swarm algorithm | |
US11911902B2 (en) | Method for obstacle avoidance in degraded environments of robots based on intrinsic plasticity of SNN | |
CN109800517B (en) | Improved reverse modeling method for magnetorheological damper | |
Xia et al. | Research on collision avoidance algorithm of unmanned surface vehicle based on deep reinforcement learning | |
Zhang et al. | Path planning method for unmanned surface vehicle based on RRT* and DWA |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |