CN108569138A - A kind of four wheel speed method for independently controlling of pure electric automobile based on neural network - Google Patents
A kind of four wheel speed method for independently controlling of pure electric automobile based on neural network Download PDFInfo
- Publication number
- CN108569138A CN108569138A CN201810460902.1A CN201810460902A CN108569138A CN 108569138 A CN108569138 A CN 108569138A CN 201810460902 A CN201810460902 A CN 201810460902A CN 108569138 A CN108569138 A CN 108569138A
- Authority
- CN
- China
- Prior art keywords
- neural network
- wheel
- hopfield
- pure electric
- electric automobile
- 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
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60K—ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES; ARRANGEMENT OR MOUNTING OF PLURAL DIVERSE PRIME-MOVERS IN VEHICLES; AUXILIARY DRIVES FOR VEHICLES; INSTRUMENTATION OR DASHBOARDS FOR VEHICLES; ARRANGEMENTS IN CONNECTION WITH COOLING, AIR INTAKE, GAS EXHAUST OR FUEL SUPPLY OF PROPULSION UNITS IN VEHICLES
- B60K17/00—Arrangement or mounting of transmissions in vehicles
- B60K17/34—Arrangement or mounting of transmissions in vehicles for driving both front and rear wheels, e.g. four wheel drive vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Arrangement And Driving Of Transmission Devices (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The present invention relates to a kind of theoretical methods using four vehicle wheel rotational speed independent control of Hopfield neural networks pure electric automobile, generally comprise following steps：1. being based on Hopfield network theories, a kind of Hopfield nets of 4 output of 4 inputs are established；2. collecting the driving intention of driver:Accelerator open degree, brake pedal plate aperture, gear, steering wheel angle and four vehicle wheel rotational speeds, establish driving data library；3. rational performance function and error threshold is arranged, using collecting 80% part of the driving data training Hopfield nets suitable network weight of acquisition come and biasing；4. re-using remaining driving data examines Hopfield nets, the practicable control network that can control four wheel speed of pure electric vehicle four-wheel drive car is obtained after fine tuning Hopfield nets.
Description
Technical field
The present invention relates to a kind of method of four wheel speed independent control of pure electric automobile, especially one kind is based on
The method of the four wheel speed independent control of pure electric automobile of Hopfield neural networks, belongs to automobile technical field.
Background technology
In recent years, with the rise of electric vehicle, due to its zero-emission or near-zero release, with environmental protection and energy problem
Of increasing concern, the research and development of electric vehicle are counted as solving a kind of effective means of energy environment issues.And in electricity
In many power-driven systems of electrical automobile, In-wheel motor driving as a kind of emerging electric vehicle drive form, just increasingly at
For the research emphasis and hot spot in electric vehicle field.Wheel hub motor is directly mounted in hub for vehicle wheel, the arrangement of drive system
Very flexibly, two front-wheels of electric vehicle, two trailing wheels or four wheels can be arranged according to the type of drive of vehicle
In wheel hub, make electric vehicle as 2 front-wheel drives, the mode of 2 rear wheel drives or 4 wheel drives.With internal-combustion engines vehicle and list
Motor centralized driving electric vehicle is compared, and wheel hub motor has more foreground.
On the 4 wheel electric vehicles using In-wheel motor driving system, if further importing wire controlled four wheel steering technology
(4WS), without the bulky machine driven system of traditional combustion engine automobile, and its is small, specific power is big, has very high
Transmission efficiency can greatly simplify whole structure and reduce complete vehicle weight and center of gravity, to reduce batteries of electric automobile consumption and carry
High electric vehicle stability.In-wheel motor driving form is known as the ultimate drive form of electric vehicle by industry.
The method of four wheel coordinate operations of existing control mainly has electronic differential and Torque Control, and nerve was not used
Network control method, under the intelligentized continuous development of future automobile, controlling running car using neural network control method is
One very promising method.
Invention content
It is an object of the invention to：In view of the defects existing in the prior art, a kind of pure electric vehicle based on neural network is proposed
Automobile four-wheel rotating speed method for independently controlling, according to the behavior of driver, using neural network control method, according to four, automobile
The actual speed of wheel distributes rational acceleration to each wheel, enables four wheel coordinate operations.
In order to reach object above, the present invention provides a kind of, and four wheel speed of pure electric automobile based on neural network is only
Vertical control method, using all driving behaviors of Hopfield neural network analysis drivers same time, and is closed on this basis
Reason distribution four-wheel speed, drives vehicle smooth-ride；
Include the following steps：
Step 1) is directed under various driving environments, the driving behavior in driver's same time, the driving of collection vehicle
Data establish situation of remote database, and thus obtain sample；
Step 2), by extracting vehicle-state feature and driving behavior in the sample data in step 1, according to sample
Feature determines the input quantity and output quantity of Hopfield neural networks, builds Hopfield neural network models：Including construction god
Object function, energy function through network and dynamical equation, the weight w between each neuronijB is inputted with biasingi；
Step 3) is input to the feature vector in step 2 as training set data in Hopfield neural network models
Parsing training is carried out, optimization Hopfield neural network models are to complete Hopfield neural network speed setting controllers；
Step 4) inputs the real time data of acquisition into trained Hopfield neural networks speed setting controller,
Matching generates the required vehicle data of driving vehicle smooth-ride.
Further, in the step 1), driving data includes by gas pedal aperture, brake pedal aperture, gear, side
The driver formed to disk corner controls the rotating speed of intent data and four wheels；And the data of collection are handled, it will be oily
Door pedal aperture, brake pedal aperture, the data of gear and steering wheel angle are processed into 4 × 1 column vector pq, and normalize；It will
The rotating speed of four wheels is processed into 4 × 1 column vector tq, and normalizes.
Further, in the step 2), the controlling behavior of vehicle is analyzed according to driver, show that it controls shadow
Ring car speed four variables be：Gas pedal aperture, brake pedal aperture, gear, steering wheel angle, therefore set
The input quantity of Hopfield neural networks is four, including gas pedal aperture, brake pedal aperture, gear and steering wheel turn
Angle；Output quantity is four, including four respective angular acceleration of wheel, to control the increase and decrease of four vehicle wheel rotational speeds.
Further, number of the Hopfield neural network models based on input quantity and output quantity, two layers of structure
Network structure, respectively feedover layer and recurrence layer；Recurrence layer is initialized using the output of feedforward layer, and standard is pointed out in output
Relationship between pattern and input vector.
Further, in the step 3, build learning rules, successively change Hopfield neural networks weights and partially
It sets；The learning rules are Wnew=Wold+ △ W, bnew=bold+△b。
The performance function f (X) and error threshold of quadratic function type are set according to learning rules, when performance function value reaches
Weights and biasing are changed when global minima, when sample mean square deviation is in allowable error threshold value, deconditioning Hopfield nerves
Network；
Using the Hebb learning rules with decaying, the connection weight matrix for obtaining model is：
W (q)=(1- γ) W (q-1)+α a (q) PT(q),
Wherein, learning rate isλmaxIt is the maximum eigenvalue of Hassian matrix A, output rotating speed aqBy acceleration
Product αi(i=1,2,3,4) divide gained；
The quadratic performance function
The error threshold is 0.01, and when sample mean square deviation is less than error threshold, training stops.
Further, in the feedforward layer, weights are 4 × 4 matrixes, are biased to 4 × 1 column vectors；In the recurrence layer, power
Value is 4 × 4 matrixes, no biasing.
Further, four wheels of the pure electric automobile are independently installed there are one wheel hub motor, and the wheel hub is electric
Machine is used to respectively drive or the corresponding wheel of feedback, and the principle drawn close to intermediate speed using high and low rotating speed, by straight
Direct torque electric current and voltage are connect with regulation motor rotating speed.
Further, in the step 4, Hopfield neural network speed setting controllers receive driver and control signal of intent
And four vehicles turn wheel speed signal, handle signal according to learning rules, are handled according to current wheel actual speed and driver intention
Signal provides correct Acceleration Control signal, and sends control signals to four wheel hub motors；Wherein, driver controls
It is intended between Hopfield neural network speed setting controllers be one-way transmission, vehicle wheel rotational speed and Hopfield neural network tune
It is transmitted in both directions between fast controller.
Beneficial effects of the present invention include：(1) control method provided through the invention can solve pure electric automobile
Each vehicle wheel rotational speed problem of disharmony during more independent drivings of wheel, coordinates four vehicles in the case where catering to driver intention
Wheel speed.(2) many mechanical driving devices are saved, using electronic control system, complete vehicle weight and center of gravity are reduced, to reduce
Batteries of electric automobile consumes and improves electric vehicle stability.
Description of the drawings
The present invention will be further described below with reference to the drawings.
Fig. 1 is the illustraton of model of the present invention.
Fig. 2 is that Hopfield neural network weights adjust flow diagram in the present invention.
Fig. 3 is Hopfield neural network neuromere learning rules schematic diagrames in the present invention.
Specific implementation mode
It present embodiments provides a kind of based on four wheel speed independent control of Hopfield neural networks pure electric automobile
Method, using all driving behaviors of Hopfield neural network analysis drivers same time, and on this basis rationally point
With four-wheel speed, vehicle smooth-ride is driven, as shown in Figure 1.Wherein, 1- driver, which controls, is intended to, gas pedal aperture, system
The signal of dynamic pedal aperture, gear and steering wheel angle all can unidirectionally be transmitted directly to Hopfield neural network speed regulating controls
Device.2- driver, which controls, to be intended between Hopfield neural network speed setting controllers be one-way transmission.3-Hopfield nerves
Network speed setting controller.Receive driver's control signal of intent and four vehicles turn wheel speed signal, and handling signal later will control
Signal is sent to four wheel hub motors.Wheel hub motor in 4- wheels.5- vehicle wheel rotational speeds and Hopfield neural network speed governing controls
It is transmitted in both directions between device processed, Hopfield handles signal according to current wheel actual speed and driver intention, provides correct
Reaction.
Include the following steps：
Step 1) is directed under various driving environments, the driving behavior in driver's same time, the driving of collection vehicle
Data include the rotating speed of gas pedal aperture, brake pedal aperture, gear, steering wheel angle and four wheels.
Vehicle carries out network communication using CAN bus, and driver is arrived driving/braking signal transmission by pedal assembly
CAN network, while onboard sensor, hub motor control device and other electrical equipments acquire signal transmission to CAN communication network
On, entire car controller acquires CAN network real-time data signal, judges according to integrated vehicle control tactics and sends command adapted thereto
To CAN communication network, and corresponding data is recorded, and establishes situation of remote database, and the data of collection are handled,
The data of gas pedal aperture, brake pedal aperture, gear and steering wheel angle are processed into 4 × 1 column vector pq, and normalizing
Change；The rotating speed of four wheels is processed into 4 × 1 column vector tq, and is normalized, sample is thus obtained.
Step 2), by extracting vehicle-state feature and driving behavior in the sample data in step 1, according to driving
Member analyzes the controlling behavior of vehicle, show that four variables that its control influences car speed are：Gas pedal aperture,
Brake pedal aperture, gear, steering wheel angle, therefore the input quantity of Hopfield neural networks is set as four, including throttle
Pedal aperture, brake pedal aperture, gear and steering wheel angle；And output quantity is then four, including four respective angles of wheel
Acceleration, to control the increase and decrease of four vehicle wheel rotational speeds.Thus Hopfield neural network models are built：Including constructing nerve net
Object function, energy function and the dynamical equation of network, the weight w between each neuronijB is inputted with biasingi。
Number of the Hopfield neural network models based on input quantity and output quantity, structure are two-tier network structure, point
Wei not feedover layer and recurrence layer；Recurrence layer using feedforward layer output initialized, output point out mode standard and input to
Relationship between amount.Wherein, it feedovers in layer, weights are 4 × 4 matrixes, are biased to 4 × 1 column vectors；In recurrence layer, weights 4
× 4 matrixes, no biasing.
Step 3) is input to the feature vector in step 2 as training set data in Hopfield neural network models
Parsing training is carried out, optimization Hopfield neural network models are to complete Hopfield neural network speed setting controllers；Structure is learned
Rule is practised, changes weights and the biasing of Hopfield neural networks successively；The learning rules are Wnew=Wold+ △ W, bnew=
bold+△b。
The performance function f (X) and error threshold of quadratic function type are set according to learning rules, when performance function value reaches
Weights and biasing are changed when global minima, when sample mean square deviation is in allowable error threshold value, deconditioning Hopfield nerves
Network；
Using the Hebb learning rules with decaying, the connection weight matrix for obtaining model is：
W (q)=(1- γ) W (q-1)+α a (q) PT(q),
Wherein, learning rate isλmaxIt is the maximum eigenvalue of Hassian matrix A, output rotating speed aqBy acceleration
Product αi(i=1,2,3,4) divide gained；
The quadratic performance function
The error threshold is 0.01, and when sample mean square deviation is less than error threshold, training stops.
Step 4) inputs the real time data of acquisition into trained Hopfield neural networks, and matching, which generates, drives
The required vehicle data of motor-car smooth-ride.The principle drawn close to intermediate speed using high and low rotating speed.In traveling, wheel hub electricity
Machine drives the torque of each wheel to be controlled and distributed by central controller, and when the vehicle is turning, central controller is according to side
Contact with road surface corner from the feedback of friction to disk and wheel, the rotating speed of foreign steamer in adjust automatically vehicle；When there is some vehicle
Wheel run out of steam skidding when, central controller according to the torque of the corresponding wheel hub motor of feedback adjustment export, carry out corresponding wheel
Braking.
In addition to the implementation, the present invention can also have other embodiment.It is all to use equivalent substitution or equivalent transformation shape
At technical solution, fall within the scope of protection required by the present invention.
Claims (10)
1. a kind of four wheel speed method for independently controlling of pure electric automobile based on neural network, it is characterised in that：It utilizes
All driving behaviors of Hopfield neural network analysis drivers same time, and reasonable distribution four-wheel speed on this basis,
Drive vehicle smooth-ride；
Include the following steps：
Step 1) is directed under various driving environments, the driving behavior in driver's same time, the driving data of collection vehicle,
Situation of remote database is established, and thus obtains sample；
Step 2), by extracting vehicle-state feature and driving behavior in the sample data in step 1, it is true according to sample characteristics
The input quantity and output quantity for determining Hopfield neural networks, build Hopfield neural network models：Including constructing neural network
Object function, energy function and dynamical equation, the weight w between each neuronijB is inputted with biasingi；
Feature vector in step 2 is input to as training set data in Hopfield neural network models and carries out by step 3)
Parsing training, optimization Hopfield neural network models are to complete Hopfield neural network speed setting controllers；
Step 4) inputs the real time data of acquisition into trained Hopfield neural networks speed setting controller, matching
Generate the required vehicle data of driving vehicle smooth-ride.
2. the pure electric automobile four wheel speed method for independently controlling according to claim 1 based on neural network, feature
It is:In the step 1), driving data includes being made of gas pedal aperture, brake pedal aperture, gear, steering wheel angle
Driver control the rotating speed of intent data and four wheels；And the data of collection are handled, by gas pedal aperture, system
The data of dynamic pedal aperture, gear and steering wheel angle are processed into 4 × 1 column vector pq, and normalize；By the rotating speed of four wheels
It is processed into 4 × 1 column vector tq, and is normalized.
3. the pure electric automobile four wheel speed method for independently controlling according to claim 1 based on neural network, feature
It is:In the step 2), the controlling behavior of vehicle is analyzed according to driver, show that its control influences car speed
Four variables are：Gas pedal aperture, brake pedal aperture, gear, steering wheel angle, therefore set Hopfield neural networks
Input quantity be four, including gas pedal aperture, brake pedal aperture, gear and steering wheel angle；
Output quantity is four, including four respective angular acceleration of wheel, to control the increase and decrease of four vehicle wheel rotational speeds.
4. the pure electric automobile four wheel speed method for independently controlling according to claim 3 based on neural network, feature
It is：Number of the Hopfield neural network models based on input quantity and output quantity, structure are two-tier network structure, point
Wei not feedover layer and recurrence layer；Recurrence layer using feedforward layer output initialized, output point out mode standard and input to
Relationship between amount.
5. the pure electric automobile four wheel speed method for independently controlling according to claim 1 based on neural network, feature
It is：In the step 3, learning rules are built, change weights and the biasing of Hopfield neural networks successively；The study rule
It is then Wnew=Wold+ △ W, bnew=bold+△b。
6. the pure electric automobile four wheel speed method for independently controlling according to claim 5 based on neural network, feature
It is：In the step 3, the performance function f (X) and error threshold of quadratic function type are set according to learning rules, when performance letter
Modification weights and biasing when numerical value reaches global minima, when sample mean square deviation is in allowable error threshold value, deconditioning
Hopfield neural networks；
Using the Hebb learning rules with decaying, the connection weight matrix for obtaining model is：
W (q)=(1- γ) W (q-1)+α a (q) PT(q),
Wherein, learning rate isλmaxIt is the maximum eigenvalue of Hassian matrix A, output rotating speed aqα is accumulated by accelerationi(i
=1,2,3,4) divide gained；
The quadratic performance function
7. the pure electric automobile four wheel speed method for independently controlling according to claim 6 based on neural network, feature
It is：The error threshold is 0.01, and when sample mean square deviation is less than error threshold, training stops.
8. the pure electric automobile four wheel speed method for independently controlling according to claim 5 based on neural network, feature
It is：In the feedforward layer, weights are 4 × 4 matrixes, are biased to 4 × 1 column vectors；In the recurrence layer, weights are 4 × 4 squares
Battle array, no biasing.
9. the pure electric automobile four wheel speed method for independently controlling according to claim 1 based on neural network, feature
It is：Four wheels of the pure electric automobile are independently installed there are one wheel hub motor, and the wheel hub motor is used to distinguish
Driving or the corresponding wheel of feedback, and the principle drawn close to intermediate speed using high and low rotating speed, pass through Direct Torque Control
Electric current and voltage are with regulation motor rotating speed.
10. the pure electric automobile four wheel speed method for independently controlling according to claim 1 based on neural network, feature
It is：In the step 4, Hopfield neural network speed setting controllers receive driver and control signal of intent and four vehicle runners
Fast signal, signal is handled according to learning rules, and signal is handled according to current wheel actual speed and driver intention, is provided correct
Acceleration Control signal, and send control signals to four wheel hub motors；Wherein, driver controls intention and Hopfield
It is one-way transmission between neural network speed setting controller, is double between vehicle wheel rotational speed and Hopfield neural network speed setting controllers
To transmission.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810460902.1A CN108569138B (en) | 2018-05-15 | 2018-05-15 | Pure electric vehicle four-wheel rotation speed independent control method based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810460902.1A CN108569138B (en) | 2018-05-15 | 2018-05-15 | Pure electric vehicle four-wheel rotation speed independent control method based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108569138A true CN108569138A (en) | 2018-09-25 |
| CN108569138B CN108569138B (en) | 2021-06-11 |
Family
ID=63572404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810460902.1A Active CN108569138B (en) | 2018-05-15 | 2018-05-15 | Pure electric vehicle four-wheel rotation speed independent control method based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108569138B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109334469A (en) * | 2018-10-23 | 2019-02-15 | 珠海格力电器股份有限公司 | A kind of method of differential control electric car vehicle wheel rotational speed |
| CN110188683A (en) * | 2019-05-30 | 2019-08-30 | 北京理工大学 | A kind of automatic Pilot control method based on CNN-LSTM |
| CN111722610A (en) * | 2019-03-20 | 2020-09-29 | 北京新能源汽车股份有限公司 | Test case generation method and device |
| CN116968704A (en) * | 2023-09-21 | 2023-10-31 | 小米汽车科技有限公司 | Vehicle brake control method, device, storage medium and vehicle |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103605285A (en) * | 2013-11-21 | 2014-02-26 | 南京理工大学 | Fuzzy nerve network control method for automobile driving robot system |
| CN104210383A (en) * | 2014-09-18 | 2014-12-17 | 上海工程技术大学 | Four-wheel independently driven electric vehicle torque distribution control method and system |
| CN105136469A (en) * | 2015-07-23 | 2015-12-09 | 江苏大学 | Unmanned vehicle speed control method based on PSO and RBF neutral network |
| CN103950471B (en) * | 2014-04-03 | 2016-03-02 | 吉林大学 | Two track unit adaptive steering system and implementation method |
| KR20180009942A (en) * | 2016-07-20 | 2018-01-30 | 현대오트론 주식회사 | Apparatus and method for deciding tire deflation |
| CN107696915A (en) * | 2017-09-20 | 2018-02-16 | 江苏大学 | A kind of wheeled driving control system of electric automobile based on hierarchical control and its control method |
-
2018
- 2018-05-15 CN CN201810460902.1A patent/CN108569138B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103605285A (en) * | 2013-11-21 | 2014-02-26 | 南京理工大学 | Fuzzy nerve network control method for automobile driving robot system |
| CN103950471B (en) * | 2014-04-03 | 2016-03-02 | 吉林大学 | Two track unit adaptive steering system and implementation method |
| CN104210383A (en) * | 2014-09-18 | 2014-12-17 | 上海工程技术大学 | Four-wheel independently driven electric vehicle torque distribution control method and system |
| CN105136469A (en) * | 2015-07-23 | 2015-12-09 | 江苏大学 | Unmanned vehicle speed control method based on PSO and RBF neutral network |
| KR20180009942A (en) * | 2016-07-20 | 2018-01-30 | 현대오트론 주식회사 | Apparatus and method for deciding tire deflation |
| CN107696915A (en) * | 2017-09-20 | 2018-02-16 | 江苏大学 | A kind of wheeled driving control system of electric automobile based on hierarchical control and its control method |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109334469A (en) * | 2018-10-23 | 2019-02-15 | 珠海格力电器股份有限公司 | A kind of method of differential control electric car vehicle wheel rotational speed |
| CN109334469B (en) * | 2018-10-23 | 2020-12-22 | 珠海格力电器股份有限公司 | Method for differentially controlling wheel rotation speed of electric automobile |
| CN111722610A (en) * | 2019-03-20 | 2020-09-29 | 北京新能源汽车股份有限公司 | Test case generation method and device |
| CN110188683A (en) * | 2019-05-30 | 2019-08-30 | 北京理工大学 | A kind of automatic Pilot control method based on CNN-LSTM |
| CN116968704A (en) * | 2023-09-21 | 2023-10-31 | 小米汽车科技有限公司 | Vehicle brake control method, device, storage medium and vehicle |
| CN116968704B (en) * | 2023-09-21 | 2024-01-02 | 小米汽车科技有限公司 | Vehicle brake control method, device, storage medium and vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108569138B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106585425B (en) | A kind of hierarchical system and control method for four hub motor driven electric vehicles | |
| CN108569138A (en) | A kind of four wheel speed method for independently controlling of pure electric automobile based on neural network | |
| CN105741637B (en) | Four-wheel hub motor electric car automated steering control method | |
| CN107719372B (en) | Four-drive electric car dynamics multi objective control system based on dynamic control allocation | |
| CN108422901B (en) | A kind of In-wheel-motor driving wheel of vehicle torque Multipurpose Optimal Method optimal based on vehicle comprehensive performance | |
| CN112116156B (en) | Hybrid train energy management method and system based on deep reinforcement learning | |
| CN102030007B (en) | Method for acquiring overall dynamics controlled quantity of independently driven-independent steering vehicle | |
| CN108482481B (en) | Four-wheel steering control method for four-wheel independent drive and steering electric automobile | |
| CN107139775A (en) | A kind of electric car direct yaw moment control method based on Non-smooth surface technology | |
| CN104210383A (en) | Four-wheel independently driven electric vehicle torque distribution control method and system | |
| CN106394560B (en) | A kind of wheel hub driving electric automobile chassis coordinated control system | |
| CN113635879B (en) | Vehicle braking force distribution method | |
| Hu et al. | An optimal torque distribution control strategy for four-wheel independent drive electric vehicles considering energy economy | |
| CN105644566B (en) | A kind of tracking of the electric automobile auxiliary lane-change track based on car networking | |
| CN111332125A (en) | Improved vehicle braking energy recovery control method and device, vehicle and storage medium | |
| CN109624732A (en) | A kind of multilayer Anti-slip regulation control method suitable for In-wheel-motor driving vehicle | |
| CN110386191A (en) | A kind of electro-hydraulic intelligent steering system performance monitor and performance optimization method | |
| Wang et al. | Vehicle lateral stability control based on stability category recognition with improved brain emotional learning network | |
| Niasar et al. | Yaw moment control via emotional adaptive neuro-fuzzy controller for independent rear wheel drives of an electric vehicle | |
| CN111332126A (en) | Vehicle braking energy recovery control method and device, vehicle and storage medium | |
| CN117021977B (en) | Method for generating driving force integrated control strategy of multi-wheel independent electric drive vehicle | |
| CN112373293A (en) | Fault processing method for distributed driving system of hub motor | |
| Zhang et al. | A fuzzy control strategy and optimization for four wheel steering system | |
| CN112810602B (en) | Intelligent line control chassis personalized stability control method and system based on clustering | |
| CN112124299B (en) | Intelligent networking new energy automobile layered energy consumption optimization method |
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 |