WO2012088635A1 - 车辆自适应巡航控制系统及方法 - Google Patents
车辆自适应巡航控制系统及方法 Download PDFInfo
- Publication number
- WO2012088635A1 WO2012088635A1 PCT/CN2010/002208 CN2010002208W WO2012088635A1 WO 2012088635 A1 WO2012088635 A1 WO 2012088635A1 CN 2010002208 W CN2010002208 W CN 2010002208W WO 2012088635 A1 WO2012088635 A1 WO 2012088635A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- vehicle
- control
- control unit
- unit
- adaptive cruise
- Prior art date
Links
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 83
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000011156 evaluation Methods 0.000 claims abstract description 75
- 230000000694 effects Effects 0.000 claims abstract description 32
- 230000013016 learning Effects 0.000 claims description 76
- 230000006870 function Effects 0.000 claims description 35
- 238000004088 simulation Methods 0.000 claims description 27
- 230000001133 acceleration Effects 0.000 claims description 26
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 230000000994 depressogenic effect Effects 0.000 claims 2
- 239000011159 matrix material Substances 0.000 claims 2
- 210000005036 nerve Anatomy 0.000 claims 1
- 238000013480 data collection Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 12
- 238000005457 optimization Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 206010039203 Road traffic accident Diseases 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012887 quadratic function Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
- B60K31/00—Vehicle fittings, acting on a single sub-unit only, for automatically controlling vehicle speed, i.e. preventing speed from exceeding an arbitrarily established velocity or maintaining speed at a particular velocity, as selected by the vehicle operator
- B60K31/0008—Vehicle fittings, acting on a single sub-unit only, for automatically controlling vehicle speed, i.e. preventing speed from exceeding an arbitrarily established velocity or maintaining speed at a particular velocity, as selected by the vehicle operator including means for detecting potential obstacles in vehicle path
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/04—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/04—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units
- B60W10/06—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units including control of combustion engines
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/18—Conjoint control of vehicle sub-units of different type or different function including control of braking systems
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/18—Conjoint control of vehicle sub-units of different type or different function including control of braking systems
- B60W10/184—Conjoint control of vehicle sub-units of different type or different function including control of braking systems with wheel brakes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/14—Adaptive cruise control
- B60W30/143—Speed control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/14—Adaptive cruise control
- B60W30/16—Control of distance between vehicles, e.g. keeping a distance to preceding vehicle
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0028—Mathematical models, e.g. for simulation
- B60W2050/0029—Mathematical model of the driver
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0062—Adapting control system settings
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2556/00—Input parameters relating to data
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2556/00—Input parameters relating to data
- B60W2556/10—Historical data
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2556/00—Input parameters relating to data
- B60W2556/45—External transmission of data to or from the vehicle
Definitions
- the present invention relates to the field of cruise control of vehicles, and more particularly to a vehicle adaptive cruise control system and method. Background technique
- the vehicle's adaptive cruise control is an advanced driver assistance system. It develops from the cruise control. By measuring the distance and relative speed between the vehicle and the preceding vehicle in real time, the appropriate throttle or brake control is calculated, and the vehicle speed control or distance control is realized by automatic adjustment.
- the adaptive cruise control effectively frees the driver from the heavy driving tasks and can realize the collision avoidance control of the vehicle, which is of great significance for improving the safety, comfort and energy saving of the driving of the vehicle.
- adaptive cruise control is a kind of assisted driving system.
- the key to its popularization is that its control effect needs to meet the characteristics of the driver. Otherwise, it will deviate from the original intention of assisted driving and is easily denied by the driver.
- Cruise control puts higher demands on it.
- Some experts question the role of adaptive cruise control and believe that it will become a technology that people have abandoned. The main reason is this, for example, in the article Rajahah B., Anceaux F., Vienne F. Trust and the use of adaptive cruise control : a study of a cut-in situation.
- the prior art vehicle adaptive cruise system has many advantages, but when controlling the vehicle speed and the vehicle distance, it cannot adaptively adjust to different environments and different driving habits to improve the user experience.
- the present invention provides an adaptive cruise control system and method.
- the adaptive cruise control system of the present invention comprises: an adaptive cruise mode selection unit for selecting different adaptive cruise modes, and differently controlling the speed and the distance of the vehicle in different modes; the data acquisition unit is used for And acquiring a state variable control unit of the vehicle, configured to generate a vehicle control variable 0 according to the state variable 0, and the evaluation unit is configured to evaluate the control effect according to the state variable ⁇ (0 and the control variable 0, if the evaluation result is control If the effect does not meet the requirements, the evaluation unit and the control unit are learned online; the throttle control unit and the brake control unit provide the unit data using the vehicle dynamics inverse model according to the control variable wo output by the control unit, and the throttle and brake are applied. Take control.
- the vehicle adaptive cruise control method of the present invention comprises the steps of: selecting an adaptive cruise mode, differently controlling the speed and the distance of the vehicle in different modes; and collecting the state of the vehicle
- the quantity x (0, the variable is used as the input of the control unit; according to the acquired state variable c (0, the vehicle control variable w is generated by the control unit (o; according to the collected vehicle state variable ⁇ (and the generated control variable ⁇ ( ⁇ ,
- the evaluation effect is evaluated by the evaluation unit. If the evaluation result is that the control effect does not meet the requirements, the evaluation unit and the control unit are performed online; if the evaluation result is that the control effect meets the requirement, the vehicle dynamics inverse model is used according to the control variable 0.
- a unit is provided to control the throttle unit and the brake unit.
- the adaptive cruise control system and method of the present invention provides an efficient construction process for human-like characteristics through off-line simulation and real vehicle experiments.
- the proposed adaptive cruise control system and method are learnable and optimistic: through offline and online learning of the driver's characteristics, it can simulate the characteristics of the driver, and the control unit is guided to guide the learning of the control unit, so that the control Features can track changes in driver characteristics.
- FIG. 1 is a block diagram showing the structure of a vehicle adaptive cruise control system of the present invention
- FIG. 2 is a schematic diagram of a three-dimensional simulation driving platform of the vehicle of the present invention.
- FIG. 3 is a schematic diagram of a humanoid adaptive cruise control process of the vehicle of the present invention.
- FIG. 4 is a flow chart of a vehicle adaptive cruise control method of the present invention. Detailed ways
- the adaptive cruise control principle of a motor vehicle is to measure the relative distance and relative speed between the vehicle and the preceding vehicle in real time through a distance sensor (usually a millimeter wave radar or a laser radar), and then calculate the vehicle speed or the safety distance according to the speed calculation of the vehicle.
- the required amount of throttle and brake control is controlled by the vehicle to achieve vehicle speed control or distance control.
- direct type adopts centralized controller, and realizes speed control or distance control directly through adjustment of throttle and brake brake
- the control task is implemented in two layers, and the upper controller calculates the desired acceleration according to the environmental information around the vehicle, realizes the vehicle speed control or the vehicle distance control, and the lower controller
- the control of the desired acceleration is achieved by adjusting the dynamics of the vehicle and by adjusting the throttle and brakes.
- the upper controller focuses on describing the driver characteristics in different driving scenarios, while the lower controller is typically implemented by establishing an inverse model of the longitudinal dynamics of the vehicle.
- the adaptive cruise control system of the present invention belongs to the above hierarchical structure, and the specific structure is as shown in FIG. 1.
- the system includes: an adaptive cruise mode selection unit 101, an adaptive cruise control unit 102, a data acquisition unit 103, and a dynamic inverse
- the adaptive cruise mode selection unit 101 is used for the driver to select an adaptive cruise mode of different modes.
- the adaptive cruise mode of the present invention can include safety, fast, and comfort, and design different performance indicator functions for each mode.
- the safety type generally means that the driving speed is relatively low, and the vehicle distance will be kept large with the preceding vehicle during driving, and the vehicle distance will be maintained when the vehicle is stopped and when the vehicle is started. If other vehicles are inserted, they will be selected. Avoiding, etc., this method is suitable for drivers who are more conservative in driving habits. This method can be used to ensure maximum safety.
- the quick type generally means that the speed is faster when it is allowed during driving, and the distance between the front and the front is usually small, and the distance is short when the vehicle stops or follows the vehicle, and the vehicle is inserted in front. It is usually not a choice to avoid, this method is suitable for groups with skilled driving experience such as taxi drivers or young people, or choose this method in case of emergency.
- the comfort is set between safe and fast, depending on the driver's driving habits. In this way, the distance maintained is a medium distance, and when other vehicles are inserted, it is judged whether it is allowed to be inserted or not accelerated by the acceleration.
- the adaptive cruise mode selection unit can be implemented by buttons, menus, joysticks, touch screens or remote controls.
- driving habits can be simplified as follows.
- d d ( d 0 + ⁇ ⁇ (t)
- a driver can select a safe type.
- ⁇ and r if it corresponds to an ordinary driver.
- the comfort type of the present invention can be selected.
- the driver of the present invention can select the quick type according to the present invention.
- the variation range of r is preferably divided into three regions, respectively corresponding to different The driver of the characteristic, for the driver to choose before driving. For example, r is less than 1 for fast, r is comfortable between 1 and 4, and r is greater than 4. Corresponding to safety. The area corresponds to the driver with different characteristics, and the driver can choose before driving.
- the automatic cruise control unit 102 is configured to generate an ideal acceleration according to the adaptive cruise mode selected by the adaptive cruise mode selection unit 101 and the data collected by the data acquisition unit 103, and then perform the acceleration by the dynamic inverse model providing unit 104.
- the solution of the brake and throttle control amounts is transmitted to the brake control unit 105 and the throttle control unit 106.
- the automatic cruise control unit 102 specifically includes an evaluation unit 1021 and a control unit 1022.
- the evaluation unit 1021 is configured to accurately estimate the performance index function R(0) of the control unit 1022 for quantitatively guiding the optimization of the control unit 1022.
- ⁇ is the return of time
- ⁇ is the discount factor
- o ⁇ ⁇ i, k is an intermediate quantity, indicating the range of time.
- p d , A and A are the adjustment weights of relative distance, relative velocity and acceleration, respectively, according to different characteristics of the driver
- ⁇ and R are also adjustment parameters, but here are replaced by 3 ⁇ 4, ⁇ ⁇ and 3 ⁇ 4, relative
- the distance (iX i) - ⁇ ( ⁇ ) is the error between the actual distance and the ideal distance ⁇ 3 ⁇ 40
- the relative speed ⁇ W v T (t) - v H (t) , which is the preceding vehicle speed (/) and the vehicle speed ( /) error. Therefore, the purpose of the optimization control is to make the performance index function R(0 max., and actually estimate the R (the calculation of dimensionality is difficult, it is difficult to calculate, and the approximation of the performance index function is used in the present invention) .
- the evaluation unit 1021 performs training through offline learning and optimizes through online learning. Offline learning is usually done before the automatic cruise system is officially used. In the offline learning process, build a 3D simulation driving platform.
- the 3D simulation driving platform is used for offline learning of the control unit and the evaluation unit by collecting driver data under different driving scenarios. Based on vehicle dynamics and 3D simulation software, the platform simulates the driving effects of adaptive cruise control.
- FIG. 2 shows a schematic diagram of a three-dimensional simulated driving platform.
- the platform includes an emulation computer 301, an animated display computer 302, a steering wheel 303, a throttle 304, a brake 305, a data collector 306, and a display 307.
- a three-dimensional display model of a vehicle, an environment, a road surface, etc. is established under a virtual reality toolbox of software such as VC++, Matlab, JAVA, etc., and three-dimensional animation parameters are set, and the animation engine of the software completes the animation, if it is passed
- the Matlab software creates an animation simulation using its xPC target output to the animation display computer 302.
- the simulation computer 301 and the animation display computer 302 communicate data by wire or wirelessly.
- the vehicle in the three-dimensional simulation driving unit is controlled by the driver's operation of the brakes and the throttle to obtain driver data under different driving scenes, which is acquired by the data collector 306 and transmitted to the simulation computer 301.
- the platform is designed with various driving scenarios, such as starting and stopping with the car, emergency braking of the front car, inserting other cars in front of the car, etc., and conducting driving experiments by skilled drivers to obtain state variables in different scenarios: 0 and control variables "( ⁇ , state variable ⁇ (0 includes relative distance and relative speed ( ⁇ (0, ⁇ 0), control variable" (0 is the acceleration of the vehicle, using these data and historical data to learn the evaluation unit.
- Time state variable ⁇ (0 can be directly through the computer program
- the sequence calculation is obtained, and the data collector 306 is responsible for receiving the control signals of the throttle and the brake, and calculating the corresponding state variable 3 ⁇ 4 according to the forward model of the dynamics of the vehicle (0, that is, the acceleration, acting on the vehicle, and according to the preceding vehicle
- the position and velocity are calculated to get the state variable ⁇ ( ⁇ +1) at the next moment.
- the platform enables joint simulation of the driver and vehicles with adaptive cruise control systems, providing a safe and fast training environment and verification method for the vehicle adaptive cruise control system.
- the hardware of the simulated driving platform includes the steering wheel, throttle, brakes, data collector, display and computer.
- the software includes a 3D simulation model of the vehicle, a simulation scene with lanes, a longitudinal dynamics model embedded in the vehicle, and an adaptive cruise control method.
- the driver implements the simulation by manipulating the driving device, which is calculated by the software and displays the results through the three-dimensional scene and data.
- the simulation verification scenarios include: cruise control without front vehicle; adaptive cruise control for acceleration of the preceding vehicle or driving, adaptive cruise control of the preceding vehicle driving away from the lane, adaptive cruise control for deceleration or parking of the preceding vehicle, Adaptive cruise control with deceleration in the case of an abnormal vehicle, adaptive cruise control with other cars inserted in front of the vehicle.
- the evaluation unit 1021 performs offline learning using the data acquired by the above platform in various driving scenarios designed, such as state variables ⁇ (0 and control variables ⁇ ).
- the evaluation unit 1021 can adopt a standard three-layer forward artificial neural network model, and fully utilizes the nonlinear function approximation ability of the artificial neural network.
- / is the integration function of the node, used to link the information from other nodes: activity or data, provide network input for this node, where the upper corner represents the number of layers;
- ⁇ is the node Activity function, used to output the activity value as the network output of this node.
- the function of each layer node is as follows.
- the integration function of the first layer node /; 1 is as follows:
- the second layer the hidden layer
- the input is weighted
- w c 2 is the hidden layer neuron weight
- exp is the exponential function
- the integration function of the second layer node /; 2 is as follows:
- the third layer the output layer, which is the weight of the output layer neuron, and the approximation of the performance index function R(i) of the output control unit "7 (0, the integration function of the third layer node / 3 is expressed as follows:
- 1 J evaluation unit 1021 learning is to adjust the learning error of the evaluation unit by adjusting the weight of the hidden layer neuron and the weight of the output layer, (0 is reduced to a predetermined threshold or the number of learning reaches a predetermined value, satisfying the artificial neural network Approaching ability.
- the evaluation unit 1021 can accurately estimate the performance index function R(0 of the adaptive cruise controller for quantitatively guiding the optimization of the control unit 1022.
- the main purpose of the evaluation unit 1021 is to adjust the connection weights of the hidden layer and the output layer, as follows:
- Dw c k dJ(t) dw c k is the learning rate of the first layer.
- an evaluation unit 1021 having a performance index requirement which may be to reduce the error to a predetermined threshold, such as a value in the range of 0.000001 to 0.1
- the indicator may also be such that the number of times of learning reaches a predetermined value, such as a value ranging from 10 to 1000000.
- the offline learning evaluation unit 1021 acquires the state variable ⁇ from the data collection unit 103, and obtains the control variable ⁇ from the control unit 1022 ( ⁇ , and calculates the evaluation index according to the state variable and the control variable). J (0, if the evaluation index meets the requirements, the control effect of the control unit 1022 is considered to be ideal, otherwise the evaluation index / / ( ⁇ ) is sent to the control unit 1022, and the control unit 1022 performs according to the received evaluation index J (0) Learn.
- Control unit 1022 is operative to generate a control signal for controlling the acceleration of the vehicle.
- Control unit 1022 is subject to offline learning and online learning processes.
- the input of the control unit 1022 is a state variable 0, and the output is a control variable (0.
- the construction process of the control unit 1022 is similar to that of the evaluation unit 1021.
- the control unit can also be constructed by using a three-layer forward artificial neural network, and the definition of each layer node is The same in the evaluation unit 1021.
- the purpose of the learning unit 1022 is to learn from the current state variable x(0, generate output "( ⁇ to maximize the performance index function, i.e., to maximize the output of the evaluation unit 1021. Therefore, the control unit 1022 is learned, and the error is defined as
- the learning process optimizes the control unit by adjusting the weights of the implicit and output layers of the neural network to meet the performance requirements and obtain a satisfactory control unit.
- the automatic cruise control unit 102 which is offline learning, performs experimental verification on the actual vehicle, and the skilled driver experiences the control effect. If the driver is not satisfied, he will turn off the automatic cruise control unit and switch to manual mode to control the vehicle. At this time, the automatic cruise control unit 102 performs online learning to bring the control effect closer to the driver's driving characteristics.
- the online learning process first performs the learning of the evaluation unit 1021, and after the evaluation unit converges, the online learning of the control unit 1022 is performed.
- Such a learning mode evaluated by the evaluation unit 1021 for the control unit 1022 can avoid the influence of the unstable operation of the driver on the performance of the control unit 1022.
- the control unit 1022 can also be learned by using the driver's control operation and the error output by the control unit 1022 as the learning error of the control unit, at which time the learning error of the control unit 1022 is defined as .
- control unit 1022 If the control unit 1022 has learned well, it can be reused as an assisted driving, and the driver chooses to switch. During the driving process, if the adaptive cruise control unit 102 still improperly handles certain driving scenarios, or the driver who replaces different characteristics is not satisfied with the characteristics of the current adaptive cruise control unit 102, it can be switched to the driver control again. The adaptive cruise control unit 102 resumes online learning.
- the control unit 1022 which is designed using an artificial neural network, is a nonlinear control method that is robust to changes in sensing parameters, such as detected relative distances and relative velocities.
- the adaptive cruise control is designed for upper layer control, and for the fluctuations due to changes in the ground friction coefficient, vehicle load changes, etc., the robustness associated with the unit 104 is provided by the underlying control, i.e., by the inverse model of vehicle dynamics.
- the above-mentioned evaluation unit 1021 and control unit 1022 can also adopt a commonly used fuzzy system side.
- the control unit 1022 can also adopt a common PID control method to obtain different learning convergence effects.
- the data acquisition unit 103 is configured to collect the relative distance between the host vehicle and the preceding vehicle, the relative speed, and the speed of the vehicle, and transmit the collected data to the adaptive cruise control unit 102.
- the data acquisition unit 103 can be a radar sensor, an ultrasonic sensor, a laser sensor, or the like.
- the unit 104 is provided according to the acceleration and dynamics inverse model outputted by the adaptive cruise control unit 102, and the control amount required for the brake control unit 105 and the throttle control unit 106 is obtained or solved to realize the control of the vehicle acceleration.
- Control input is defined as state variable x (0, including relative distance and relative speed
- the adaptive cruise control unit outputs a control variable w(i), that is, the acceleration of the vehicle, acting on the simulated driving system or the actual vehicle, and generating a state variable ⁇ ( ⁇ +1) at the next moment.
- the control effect at each moment can be given by the return r (0 in the lower right corner), and the entire performance index function is calculated by the evaluation unit in the upper right corner. (If the driver in the upper left corner thinks that the control effect is not satisfactory, it switches to driver control.
- the driver's control amount and the output of the control unit constitute an online learning error of the control unit.
- variable w of the actual vehicle is also generated by weighting (0, another learning error of the upper right corner construction control unit) (0, learning error of the right structure evaluation unit ⁇ ⁇
- the solid line in the figure indicates the direction of the data flow, the broken line indicates the direction in which the error is learned, and ⁇ 1 is the ⁇ transformation symbol, and the variable at the current time is converted to the previous time.
- the invention also provides an adaptive cruise control method for a vehicle, the method comprising the steps of: Step S401: Perform offline learning of the adaptive cruise control of the vehicle by using a three-dimensional simulation platform. This process has been described in detail earlier.
- Step S402 selecting a vehicle adaptive cruise mode.
- the adaptive cruise mode of the present invention can include safety, fast, and comfort, and design different performance indicator functions for each mode. This has been described in detail above. This mode can be selected by the driver via buttons, menus, joysticks, touch screens or remote controls.
- Step S403 Collect a state variable of the vehicle.
- the vehicle's state variable ⁇ ( ⁇ , including the relative distance and the relative speed ( ⁇ /) is collected by the distance sensor (millimeter wave radar or laser radar, etc.) carried by the vehicle during the running of the vehicle. (0, ⁇ ( ⁇ )).
- Step S404 calculating a control variable w(0, ie, a vehicle according to the collected vehicle state variable jc(i) According to the acceleration, the vehicle's dynamic inverse model is used to control the speed of the vehicle through brake control or throttle control, and the state variable ⁇ ( ⁇ +1) at the next moment is obtained.
- Step S405 the control effect of step S403 is evaluated.
- the control effect at each moment can be rewarded by the return r (given, the performance indicator function is the sum of the returns for each moment,
- p d , /3 ⁇ 4 and 3 ⁇ 4 are the adjustment weights of relative distance, relative speed and acceleration, respectively, and are selected according to different characteristics of the driver.
- the relative distance is the difference between the actual distance and the ideal distance.
- step S406 if the control effect meets the requirements, step 407 is performed to continue the adaptive cruise control. Otherwise, step 408 is performed to perform online learning of the evaluation unit and the control unit.
- the control effect can be judged by. If the driver is not satisfied with the control effect, even if it meets the requirements, it can be switched to the driver's operation by stepping on the accelerator or brake, so that the adaptive cruise control of the vehicle enters the online learning process of the evaluation unit and the control unit.
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Description
车辆自适应巡航控制系统及方法
技术领域
本发明涉及车辆的巡航控制领域, 特别涉及一种车辆自适应巡航控 制系统及方法。 背景技术
车辆的自适应巡航控制是一种先进的辅助驾驶系统。其从定速巡航控 制发展而来,通过实时测量本车与前车的距离和相对速度, 计算出合适的 油门或刹车制动的控制量, 通过自动调节实现车辆的车速控制或车距控 制。 自适应巡航控制有效地将驾驶员从繁重的驾驶任务中解脱出来, 又可 以实现车辆的避碰控制, 对于提高车辆驾驶的安全性、舒适性、和节能性 具有重要意义。
安全性。 据世界健康组织的统计结果 (参见 Broggi A., Zelinsky A., Parent M., Thorpe C. E. Intelligent vehicles, In Spring Handbooks of Robotics Siciliano B., Khatib O. (Eds.), Springer- Verlag Berlin Heidelberg 2008, pp. 1175-1198. ) , 全世界范围内每年约有 120万人死于交通事故, 5000万人 受伤,其中 90%是由于驾驶员的失误造成,包括疲劳驾驶、醉驾、超速等。 正常驾驶员从发现情况到做出反应的平均时间为 1秒,而车辆的自适应巡 航控制周期要短, 因此能有效地避免绝大多数交通事故的发生。
舒适性。 市区交通拥挤, 车辆驾驶经常起起停停, 驾驶员需要完成大 量的换档和踩离合器的工作(手动档), 大约每分钟完成 20〜30个手脚协 调动作(甘志梅. 基于激光雷达的 Stop&Go巡航控制技术研究. 上海交通 大学硕士学位论文, 2009.), 是产生驾驶员疲劳的主要原因, 而自适应巡 航控制则能将驾驶员从这种繁重劳动中完全解脱出来,使车辆驾驶真正成 为一种享受和乐趣。
节能性。近来我们倡导节能、低碳, 而低速行车时产生最多的尾气排 放, 自适应巡航控制则提供了一种优化节能的控制技术。 另一方面, 装备 了自适应巡航控制的车辆之间保持着合适的间距,有效地提高了道路的通 行能力、 缓解了交通拥挤, 具有很好的经济性。 最近研究表明, 高速公路
上装有自适应巡航控制系统的车辆比例若达到 25%,能完全消除高速公路 的拥堵现象, 在文章 Kesting A., Treiber M., Schonhof M., Helbing D.
Adaptive cruise control design for active congestion avoidance. Transportation
Research Part C, 2008, 16: 668-683.中有具体描述。
然而, 自适应巡航控制作为一种辅助驾驶系统, 其应用普及的关键在 于其控制效果需要符合驾驶员的特性,否则就背离了辅助驾驶的初衷而容 易被驾驶员否定,这就对开发自适应巡航控制提出了更高的要求。一些专 家质疑自适应巡航控制的作用, 认为其将成为人们摒弃的一项技术, 其主 要原因就在于此, 例如在文章 Rajaonah B., Anceaux F., Vienne F. Trust and the use of adaptive cruise control: a study of a cut-in situation. Cognitive
Technology Work, 2006, 8: 146-155.中对这种情况就有阐述。。
综上, 现有技术中的车辆自适应巡航系统存在诸多优点, 但在控制车 辆速度和车距时,却不能针对不同环境和不同驾驶习惯来进行自适应调节 以改善用户体验。
因此,现有的车辆自适应巡航系统的功能亟待改善, 以使其更好地发 挥辅助驾驶的作用。 发明内容
为了解决现有技术存在的上述问题,本发明提供了一种自适应巡航控 制系统和方法。
本发明的自适应巡航控制系统包括: 自适应巡航模式选择单元,用于 选择不同的自适应巡航模式,在不同的模式下对车辆的速度与车距进行不 同的控制; 数据采集单元, 用于采集车辆的状态变量 控制单元, 用 于根据所述状态变量 0生成车辆控制变量 0, 评价单元, 用于根据所 述状态变量 χ(0和控制变量 0对控制效果进行评价, 如果评价结果为控 制效果不符合要求, 则使评价单元和控制单元进行在线学习; 油门控制单 元和制动控制单元, 根据控制单元输出的控制变量 wo, 利用车辆动力学 逆模型提供单元的数据, 对油门和制动进行控制。
本发明的车辆自适应巡航控制方法包括步骤: 选择自适应巡航模式, 在不同的模式下对车辆的速度与车距进行不同的控制;采集车辆的状态变
量 x(0, 该变量作为控制单元的输入; 根据采集的状态变量 c(0, 由控制 单元生成车辆控制变量 w(o; 根据采集的车辆状态变量 χ( 和生成的控制 变量 Μ(ο, 由评价单元对控制效果进行评价, 如果评价结果为控制效果不 符合要求, 则进行评价单元和控制单元在线学习; 如果评价结果为控制效 果符合要求, 则根据控制变量 0, 利用车辆动力学逆模型提供单元, 对 油门单元和制动单元进行控制。
本发明的自适应巡航控制系统和方法, 通过离线仿真和实车实验, 提 供了一种有效的仿人特性的构造过程。所提出的自适应巡航控制系统和方 法具有学习性和优化性: 通过对驾驶员特性的离线和在线学习, 使其能模 仿驾驶员的特性, 而且通过构造评价单元指导控制单元的学习, 使控制特 性能跟踪驾驶员特性的变化。 附图说明
图 1是本发明的车辆自适应巡航控制系统的结构框图;
图 2是本发明车辆的三维仿真驾驶平台示意图;
图 3是本发明车辆的仿人式自适应巡航控制过程示意图;
图 4是本发明的车辆自适应巡航控制方法的流程图。 具体实施方式
下面结合附图详细说明本发明技术方案中所涉及的各个细节问题。应 指出的是, 所描述的实施例仅旨在便于对本发明的理解, 而对其不起任何 限定作用。
机动车的自适应巡航控制原理是通过距离传感器 (通常为毫米波雷达 或激光雷达)实时测量本车与前车的相对距离、 相对速度, 然后结合本车 速度解算出实现车速或安全车距保持所需要的油门和刹车制动的控制量, 通过自动调节实现车辆的车速控制或车距控制。
目前所提出的自适应巡航控制的结构大致分为直接式和分层式两种: 直接式采用集中控制器,直接通过对油门和刹车制动的调节实现车速控制 或车距控制; 分层式则将控制任务分两层实现, 上层控制器根据本车周围 的环境信息计算期望的加速度、实现车速控制或车距控制, 而下层控制器
则考虑车辆的动力学特性、通过对油门和刹车制动的调节实现对期望加速 度的控制。上层控制器着重于描述在不同驾驶场景下的驾驶员特性, 而下 层控制器通常通过建立车辆纵向动力学的逆模型来实现。
本发明的自适应巡航控制系统属于上述的分层式, 具体结构如图 1 所示, 该系统包括: 自适应巡航模式选择单元 101、 自适应巡航控制单元 102、 数据采集单元 103、 动力学逆模型提供单元 104、 制动控制单元 105 以及油门控制单元 106。
自适应巡航模式选择单元 101 用于供驾驶员选择不同模式的自适应 巡航模式。
驾驶员的驾驶习惯有很大差别, 与职业、 性格、 性别、 年龄等有关, 如出租车司机和新手, 年轻人和老年人的驾驶习惯区别很大。 因此, 需要 设计符合各种驾驶习惯的自适应巡航控制方法,该辅助驾驶系统才能广为 接受。
本发明的自适应巡航模式可以包括安全式、快速式和舒适式, 并为每 种方式设计不同的性能指标函数。
安全式一般是指行驶速度比较低,在驾驶过程中会与前车保持大的车 距, 跟车停止时与跟车起动时都会保持大的车距, 如果有其它车辆插入的 时候, 都会选择避让等等, 这种方式适合驾驶习惯比较保守的驾驶员, 选 择这种方式可最大程度的保证安全性。
快速式一般是指在驾驶过程中在允许的情况下车速较快,通常与前车 的车距很小, 在跟车停止或跟车启动时保持很短的车距, 在前方有车插入 的时候通常不会选择避让,这种方式适合于例如出租车司机或年轻人等有 熟练驾驶经验的群体, 或者在有紧急事情的情况下选择这种方式。
舒适式设置在安全式与快速式之间,可根据驾驶员的驾驶习惯来进行 选择。在这种方式下, 保持的车距是中等车距, 在有其它车辆插入时会根 据情况来判断是允许插入还是通过加速不允许插入。
自适应巡航模式选择单元可通过按钮、 菜单、操作杆、触摸屏或遥控 器等来实现。
在现有技术中, 对驾驶习惯的研究分析已有很多工作, 如通过对驾驶 数据的采集, 可以简化地将驾驶习惯表示为下式
dd ( = d0 + τντ (t) 式中, 为停车时本车和前车间的距离, ( )为行车时 ί时刻本车和前 车的理想间距, 为前车速度, r定义为一个与驾驶员特性相关的线性 系数。 和 r若比较大, 则对应比较保守的驾驶员, 在本发明中, 这样的驾 驶员可以选择安全式。 ^和 r若中, 则对应普通驾驶员, 可以选择本发明 的舒适式。 。和 r若小, 则对应比较性急的驾驶员, 可以选择本发明的快 速式。 在本发明, 优选地将 r的变化范围分为三个区域, 分别对应不同特 性的驾驶员, 供驾驶员在开车前进行选择。 例如, r小于 1对应快速式, r在 1与 4之间对应舒适式, r大于 4对应安全式。可根据实际需要调整 r 的三个区域, 分别对应不同特性的驾驶员, 供驾驶员在开车前进行选择。
自动巡航控制单元 102用于根据自适应巡航模式选择单元 101选择的 自适应巡航模式以及数据采集单元 103采集的数据,经过处理产生理想的 加速度,然后将该加速度由动力学逆模型提供单元 104进行制动和油门控 制量的解算, 传送到制动控制单元 105和油门控制单元 106。
自动巡航控制单元 102具体地包括评价单元 1021和控制单元 1022。 评价单元 1021用于准确估计出控制单元 1022的性能指标函数 R(0, 用来定量指导控制单元 1022的优化。
定义性能指标函数 R( 为每个时刻回报 的累加和, 以指导控制单 元的优化过程,
式中, ο为 时刻的回报, ζ为折扣因子, o < ≤i, k是一个中间量, 表示时刻 的范围。 可以用规范的二次型函数定义回报 为:
r{t) = ~{x(t)T Qx(t) + u(tf Ru(t)) = -{pdM{t)2 +
式中, pd, A和 A分别为相对距离、 相对速度和加速度的调节权重, 根据 驾驶员的不同特性进行选取, ρ和 R也是调节参数, 但这里由 ¾, ρν和 ¾ 取代,相对距离 (iX i) - ^ (ί),为实际距离 与理想距离 ί¾0的误 差,相对速度 Δν W = vT (t) - vH (t) ,为前车速度 (/)和本车速度 (/)的误 差。 因此, 优化控制的目的是使性能指标函数 R(0最大。 而实际上对 R(0 的计算带来维数灾的难题, 很难计算, 在本发明中采用性能指标函数的近 似值 )来估计。
不同的自适应巡航模式对应不同的性能指标函数的近似值 J(t)。
评价单元 1021是通过离线学习进行培训,并通过在线学习进行优化。 离线学习通常是在自动巡航系统正式使用之前完成。在离线学习过程 中, 搭建三维仿真驾驶平台。三维仿真驾驶平台通过采集不同驾驶场景下 的驾驶员数据, 用于控制单元和评价单元的离线学习。该平台基于车辆的 动力学和三维仿真软件, 能模拟出自适应巡航控制的驾驶效果。
图 2示出了三维仿真驾驶平台的示意图。该平台包括仿真计算机 301、 动画显示计算机 302、 方向盘 303、 油门 304、 刹车 305、 数据采集器 306 以及显示器 307。
在仿真计算机 301中, 利用 VC++、 Matlab、 JAVA等软件的虚拟现实 工具箱下建立车辆、 环境, 路面等的三维显示模型, 设置三维动画参数, 由该软件的动画引擎来完成动画, 如果是通过 Matlab软件建立的, 则利 用其 xPC target输出到动画显示计算机 302实现动画仿真。仿真计算机 301 和动画显示计算机 302之间通过有线或无线方式进行数据通讯。由驾驶员 对刹车和油门的操作控制三维仿真驾驶单元中的车辆,以获得不同驾驶场 景下的驾驶员数据, 该数据通过数据采集器 306获取, 并传送到仿真计算 机 301。
该平台设计各种驾驶场景, 如跟车起动和停止、前车紧急刹车、有其 他车在本车前插入等, 进行熟练驾驶员的驾驶实验, 获取不同场景下的状 态变量: 0和控制变量 "(ή, 状态变量 χ(0包括相对距离和相对速度 (Δί (0,Δ 0), 控制变量 "(0即本车的加速度, 利用这些数据和历史数据 对评价单元进行学习。 在离线学习时状态变量 χ(0可以直接通过计算机程
序计算获得, 数据采集器 306负责接收油门和刹车的控制信号, 根据车辆 的动力学的正向模型计算得到相应的状态变量 ¾(0, 也就是加速度, 作用 在本车上, 并根据前车的位置和速度计算得到下一个时刻的状态变量 χ(ί+1)。
该平台可以实现驾驶员和带有自适应巡航控制系统车辆的联合仿真, 为车辆自适应巡航控制系统提供一种安全、 快速的训练环境和验证方式。 仿真驾驶平台的硬件包括方向盘, 油门, 刹车, 数据采集器, 显示器和计 算机等。软件包括车辆的三维仿真模型, 带有车道的仿真场景, 嵌入车辆 的纵向动力学模型和自适应巡航控制方法。驾驶员通过操控驾驶装置实现 仿真, 由软件进行计算并通过三维场景和数据显示结果。进行仿真验证场 景包括: 无前车的巡航控制; 前车起动或行驶中加速的自适应巡航控制, 前车驶离车道的自适应巡航控制,前车行驶中减速或停车的自适应巡航控 制, 前车异常情况下减速的自适应巡航控制, 其他车在本车前插入的自适 应巡航控制等。
评价单元 1021利用上述平台在所设计的各种驾驶场景下所获取的数 据, 如状态变量 χ(0和控制变量 Μ , 进行离线学习。
评价单元 1021可采用标准三层前向人工神经网络模型, 充分利用人 工神经网络的非线性函数逼近能力。 为了说明各层节点函数作以下定义: /为节点的整合函数, 用以联结从其它节点来的信息: 活性或数据, 为此 节点提供网络输入, 其中上角标 代表层数; ^为节点的活性函数, 用于 输出活性值作为此节点的网络输出。 每层节点的函数功能如下,
第一层: 输入层, 起传输数据到下一层的作用, 输入变量 ^(0包括状 态变量 χ,(0, ζ'=1,2,...,ρ,和控制变量 (/),y(0=O( , (0), =1,2,...,^=/?+1), p表示状态变量的个数。 第一层节点的整合函数/; 1如下表示:
且 w
第三层: 输出层, 为输出层神经元权重, 输出控制单元的性能指 标函数 R(i)的近似值《7(0, 第三层节点的整合函数 /3如下表示:
/3=∑«且03=«/(0 = /3
1 J 评价单元 1021 的学习就是通过对隐含层神经元权重 和输出层神 经元权重 的调节, 使评价单元的学习误差 (0减小到预定阈值或学习 次数达到预定值, 满足人工神经网络的逼近能力。
利用 J( 、 《/(ί-ι)和回报 构造出评价单元的学习误差 (ο。
上式与前面的基于动态规划的性能指标函数 R(0的定义相同。 因此, 评价单元 1021 能够准确估计出自适应巡航控制器的性能指标函数 R(0, 用来定量指导控制单元 1022的优化。评价单元 1021学习的主要目的是调 整隐含层和输出层的联结权重, 具体如下:
wc k = wc k + Awc k : =!:(t) [- β]
dEc(t) _ dEc{t) dJ{t)
dwc k dJ(t) dwc k 是第 层的学习率。 利用评价单元 1021的学习误差 (0对评价单元 1021进行学习,获得 具有满足性能指标要求的评价单元 1021, 该指标可以是使误差减小到预 定阈值, 如可在 0.000001到 0.1范围内取值, 该指标也可以是使学习次数 达到预定值, 如可在 10到 1000000范围内取值。
车辆开始在路面实际行驶时, 经过离线学习的评价单元 1021从数据 采集单元 103获取状态变量 χ( ,从控制单元 1022获取控制变量 Μ(ή,并 根据所述状态变量和控制变量计算得到评价指标 J(0, 如果评价指标 符合要求, 则认为控制单元 1022的控制效果是理想的, 否则将评价指标 •/(ί)发送给控制单元 1022, 控制单元 1022根据接收到的评价指标 J(0进行 学习。
控制单元 1022用于产生控制信号, 该控制信号用于控制车辆的加速 度。 控制单元 1022需要经过离线学习和在线学习过程。
控制单元 1022的输入为状态变量 0, 输出为控制变量 (0。 控制单 元 1022的构造过程与评价单元 1021类似,同样可以采用三层前向人工神 经网络构造出控制单元, 每层节点的定义与评价单元 1021中的相同。
控制单元 1022学习的目的是根据当前状态变量 x(0、 产生输出 "(ϋ 能使性能指标函数最大, 即, 使评价单元 1021的输出 最大。 因此, 对 控制单元 1022进行学习, 误差 定义为
ea(t) = J(t) - Uc(t)
Ea{t) =→]{t) 式中 定义为效用函数, 通常设定为控制单元 1022的性能指标函数的 近似值 J(0能够接近的一个值。 在回报 r(0的最小值为零的情况下, 可以 设定效用函数 (0也为零。 学习过程通过调节神经网络隐含层和输出层 的权重, 对控制单元进行优化, 满足性能指标要求, 得到满意的控制单元
对离线学习好的自动巡航控制单元 102在实际车辆上进行实验验证, 由熟练驾驶员体验控制效果。 若驾驶员不满意他会关闭自动巡航控制单 元, 切换到人工方式对车辆进行控制。此时对自动巡航控制单元 102进行 在线学习, 使控制效果更接近驾驶员的驾驶特性。
在线学习过程时先进行评价单元 1021 的学习, 待评价单元收敛后, 再进行控制单元 1022 的在线学习。 这种通过评价单元 1021对控制单元 1022评估的学习方式, 可以避免驾驶员的不稳定操作对控制单元 1022性 能的影响。 也可以采用驾驶员的控制操作与控制单元 1022输出的误差作 为控制单元的学习误差, 对控制单元 1022进行学习, 此时控制单元 1022 的学习误差定义为。
ea(t) = a{t) - ud {t) 式中 Wa( 为控制单元 1022计算的加速度, iU0为驾驶员控制车辆的加速 度。 若在线优化后的控制单元 1022的输出与驾驶员相近, 则提示在线学 习完毕。 所产生控制车辆的加速度 M(0也可以是上述两个加速度的加权和 ua{t) = wua{t) + {\ - w)ud{t) 式中, 0=<ν <=1 为权重, 随着学习过程的进行其值逐渐增大, 直至最终 为 1, 车辆的加速度完全由控制单元 1022的输出决定。
若控制单元 1022已学习好, 可以重新用作辅助驾驶, 由驾驶员选择 切换。在驾驶过程中, 若自适应巡航控制单元 102仍然对某些驾驶场景处 理不当,或更换不同特性的驾驶员对当前自适应巡航控制单元 102的特性 不满意, 均可以再次切换为驾驶员控制, 自适应巡航控制单元 102重新进 行在线学习。
采用人工神经网络设计的控制单元 1022是一种非线性控制方法, 对 于传感参数的变化, 如检测的相对距离和相对速度具有很好的鲁棒性。所 设计的自适应巡航控制是上层控制, 对于由于地面摩擦系数变化、车辆负 载变化等的波动, 要通过下层控制、 即由车辆动力学的逆模型提供单元 104相关的鲁棒性解决。
上述的评价单元 1021和控制单元 1022也可以采用常用的模糊系统方
法, 控制单元 1022也可以采用常用的 PID控制方法, 得到不同的学习收 敛效果。数据采集单元 103用于采集本车与前车的之间的相对距离、相对 速度以及自身的速度, 并将采集的数据发送到自适应巡航控制单元 102。 该数据采集单元 103可以是雷达传感器、 超声传感器、 激光传感器等。
根据自适应巡航控制单元 102 输出的加速度和动力学逆模型提供单 元 104, 査表得到或解算出制动控制单元 105和油门控制单元 106所需要 的控制量, 实现车辆加速度的控制。
综上,结合图 3进一步理解本发明自适应巡航控制系统的离线和在线 学习原理。 控制输入定义为状态变量 x(0, 包括相对距离和相对速度
, 自适应巡航控制单元输出控制变量 w(i), 即本车的加速度, 作用在仿真驾驶系统或实际车辆上, 产生下一个时刻的状态变量 χ(ί+1)。 每一个时刻的控制效果可由右下角的回报 r(0给出, 由右上角的评价单元 计算得到整个性能指标函数 ( 。左上角的驾驶员若认为控制效果不满意, 则切换为驾驶员控制,驾驶员的控制量与控制单元的输出量构成控制单元 的一种在线学习误差 。(0,也通过加权方式生成实际车辆的控制变量 w(0, 右上角构造控制单元的另一种学习误差 £。(0,右侧构造评价单元的学习误 差 Ε ήο 图中实线表示数据流方向, 虚线表示根据误差进行学习的方向, Ζ 1为 Ζ变换符号, 将当前时刻的变量变换为前一时刻的变量
本发明还提供了一种车辆的自适应巡航控制方法, 该方法包括步骤: 步骤 S401 , 利用三维仿真平台对车辆的自适应巡航控制进行离线学 习。 该过程前面已经了详细的描述。
步骤 S402, 选择车辆自适应巡航模式。
本发明的自适应巡航模式可以包括安全式、快速式和舒适式, 并为每 种方式设计不同的性能指标函数。这在前面都已经进行了详细描述。该模 式可由驾驶员通过按钮、 菜单、 操作杆、 触摸屏或遥控器等来进行选择。
步骤 S403 , 采集车辆的状态变量。
根据本发明的车辆自适应巡航控制方法, 在车辆行驶过程中,通过车 辆携带的距离传感器 (毫米波雷达或激光雷达等) 采集车辆的状态变量 χ(ή, 包括相对距离和相对速度 (Δί/(0,Δν(ί))。
步骤 S404 , 根据所采集的车辆状态变量 jc(i)计算控制变量 w(0, 即车
辆的加速度, 根据所述加速度, 利用车辆的动力学逆模型, 通过制动控制 或油门控制来控制车辆的速度, 得到下一个时刻的状态变量 χ(ί+1)。
步骤 S405, 对步骤 S403的控制效果进行评价。
每一时刻的控制效果可通过回报 r( 给出, 性能指标函数 为每个 时刻回报的累加和,
式中, pd, /¾和¾分别为相对距离、 相对速度和加速度的调节权重, 根据 驾驶员的不同特性进行选取。 相对距离 为实际距离与理想距离的差
M {ή = d(t) - dd {ή, 相对速度 Δν W为前车速度 (t)和本车速度 (t)差
计算得到性能指标函数 R(o之后, 取其近似值 替代, 由评价单 元提供。
步骤 S406, 如果控制效果符合要求, 则执行步骤 407, 继续进行自适 应巡航控制,否则,执行步骤 408,进行评价单元和控制单元的在线学习。
在该步骤中, 控制效果是否符合要求可通过《7(0来判断, 如果 J(0满 足一定的阈值范围, 则说明控制效果良好, 可以继续进行控制, 如果 J(0 不符合要求, 则进行评价单元和控制单元的学习。
在本发明的方法中, 可通过 来判断控制效果。 如果驾驶员不满意 控制效果, 即使 符合要求, 也可以通过踩油门或刹车的方式来切换为 由驾驶员操作,从而使车辆的自适应巡航控制进入评价单元和控制单元的 在线学习过程。
在本发明的方法中, 不仅对自适应巡航控制进行在线学习,还可以进 行评价单元和控制单元的离线学习。
其中评价单元和控制单元的离线、在线学习在前面已经进行了详细描
述, 在此不再赘述。
前面己经具体描述了本发明的实施方案, 应当理解,对于一个具有本 技术领域的普通技能的人, 在不脱离本发明的范围的任何修改或局部替 换, 均属于本发明权利要求书保护的范围。
Claims
1、 一种车辆的自适应巡航控制系统, 该系统包括- 自适应巡航模式选择单元, 用于选择不同的自适应巡航模式, 在不同 的模式下对车辆的速度与车距进行不同的控制;
数据采集单元, 用于采集车辆的状态变量
控制单元, 用于根据所述状态变量 χ( 生成车辆控制变量
评价单元, 用于根据所述状态变量 χ(0和控制变量 0对控制效果进 行评价, 如果评价结果为控制效果不符合要求, 则使评价单元和控制单元 进行在线学习;
油门控制单元和制动控制单元, 根据控制单元输出的控制变量 利用车辆动力学逆模型对油门和制动进行控制。
2、 根据权利要求 1所述的系统, 其特征在于, 所述自适应巡航模式 包括安全式、 快速式和舒适式。
3、 根据权利要求 2所述的系统, 其特征在于, 所述自适应巡航模式 的选择方式包括: 按钮、 菜单、 操作杆、 触摸屏或遥控器。
4、 根据权利要求 3所述的系统, 其特征在于, 所述数据采集单元为 距离传感器或速度传感器, 所述状态变量 χ(0包括相对距离和相对速度 (Δ^( ,Δν( ), 所述控制单元生成的控制变量 w(0为车辆的加速度。
5、 根据权利要求 1-4任一项所述的系统, 其特征在于, 评价单元的 输出为性能指标函数的近^ 直,根据性能指标函数的近似值来判断控制单 元的控制效果是否符合要求, 如果符合要求, 则控制单元继续进行控制, 否则评价单元和控制单元开始在线学习;或者当车辆的油门或者刹车被踩 下时,车辆被切换为驾驶员操作方式,评价单元和控制单元开始在线学习。
6、 根据权利要求 5所述的系统, 其特征在于, 所述评价单元的在线 学习是采用标准三层前向人工神经网络模型,利用评价单元的学习误差对 评价单元进行学习,使评价单元的学习误差减小到预定值或者使学习次数 达到预定值。
7、 根据权利要求 5所述的系统, 其特征在于, 所述控制单元的在线 学习是在评价单元的在线学习收敛之后进行,采用标准三层前向人工神经 网络模型,利用控制单元的学习误差对控制单元进行学习, 使控制单元的 学习误差减小到预定值或者使学习次数达到预定值。
9、 根据权利要求 8所述的系统, 其特征在于, 评价单元的学习误差 表示为: £c(i) «7( — 1) + (),
10、根据权利要求 6或 7所述的系统, 其特征在于, 在车辆被实际驾 驶之前,评价单元和控制单元在三维仿真驾驶平台上进行与在线学习方式 相同的离线学习,三维仿真驾驶平台能够实现驾驶员和带有自适应巡航控 制系统车辆的联合仿真, 该平台的硬件包括方向盘, 油门, 刹车, 数据采 集器, 显示器和仿真计算机, 软件包括车辆的三维仿真模型, 带有车道的 仿真场景, 嵌入车辆的纵向动力学模型和自适应巡航控制方法。
11、 一种车辆的自适应巡航控制方法, 该方法包括步骤- 选择自适应巡航模式,在不同的模式下对车辆的速度与车距进行不同 的控制;
采集车辆的状态变量 x(0, 该变量被用于控制车辆的速度和车距; 根据采集的状态变量 x(0生成车辆控制变量 0;
根据采集的车辆状态变量 χ(0和生成的控制变量 o对控制效果进行 评价, 如果评价结果为控制效果不符合要求, 则进行评价单元和控制单元 在线学习;
如果评价结果为控制效果符合要求, 则根据控制变量 利用车辆 动力学逆模型对油门单元和制动单元进行控制。
12、 根据权利要求 11所述的方法, 其特征在于, 所述自适应巡航模 式包括安全式、 快速式和舒适式。
13、 根据权利要求 12所述的方法, 其特征在于, 所述自适应巡航模 式的选择方式包括: 按钮、 菜单、 操作杆、 触摸屏或遥控器。
14、 根据权利要求 13所述的方法, 其特征在于, 利用距离传感器进 行数据采集, 所述状态变量:c(0包括距离和速度误差 (Δί/(0,Δν(0), 所述 控制单元生成的控制变量 w(0为车辆的加速度。
15、 根据权利要求 11-14任一项所述的方法, 其特征在于, 评价输出 为性能指标函数的近似值,根据性能指标函数的近似值值来判断控制效果 是否符合要求, 如果符合要求, 则继续进行控制, 否则进行评价单元和控 制单元的在线学习; 或者当车辆的油门或者刹车被踩下时, 开始评价单元 和控制单元的在线学习。
16、 根据权利要求 15所述的方法, 其特征在于, 所述评价单元的在 线学习是采用标准三层前向人工神经网络模型,利用评价单元的学习误差 对评价单元进行学习,使评价单元的学习误差减小到预定值或者使学习次 数达到预定值。
17、 根据权利要求 15所述的方法, 其特征在于, 所述控制单元的在 线学习是在评价单元的在线学习收敛之后进行,采用标准三层前向人工神 经网络模型, 利用控制单元的学习误差对控制进行学习, 使控制单元的学 习误差减小到预定值或者使学习次数达到预定值。
18、根据权利要求 16或 17所述的方法, 其特征在于, 所述性能函数 指标的近似值表示为:
J{t) = X rMr(k) , 其中, y为折扣因子, 0 < y≤l,
k=t+\ r{t) 为 t 时 亥 lj 的 回 报 , 表 示 为 : r(t) = -(x{tf Qx{t) + u(tf Rw( ) = -(A + ΡΜή2 + Z ^)2),其中 pd, ρ^Πρ。分别为相对距离、 相对速度和加速度的调节权重, ρ和 R为 调节参数, Τ表示转置矩阵, 相对距离 Δί/(ί)为实际距离与理想距离的差, 相对速度 Δν(ί)为前车速度和本车速度差, (0为本车的加速度。
19、 根据权利要求 18所述的方法, 其特征在于, 评价单元的学习误 差 表示为: (ί) = (ί), 其中, ec(i) = « W-J( — 1) + 0, 控制单元的学习误差表示为: 。(ί) = (0, 其中 ^(O-JW- [/ ), 式中 (0设定为控制单元的性能指标函数的近似值 能够接近的一 个值。
20、 根据权利要求 16或 17所述的方法, 其特征在于, 在车辆被实际 驾驶之前,评价单元和控制单元在三维仿真驾驶平台上进行与在线学习方 式相同的离线学习,三维仿真驾驶平台能够实现驾驶员和带有自适应巡航 控制系统车辆的联合仿真, 硬件包括方向盘, 油门, 刹车, 数据采集器, 显示器和仿真计算机, 软件包括车辆的三维仿真模型, 带有车道的仿真场 景, 嵌入车辆的纵向动力学模型和自适应巡航控制方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/976,614 US9266533B2 (en) | 2010-12-30 | 2010-12-30 | Adaptive cruise control system and method for vehicle |
PCT/CN2010/002208 WO2012088635A1 (zh) | 2010-12-30 | 2010-12-30 | 车辆自适应巡航控制系统及方法 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2010/002208 WO2012088635A1 (zh) | 2010-12-30 | 2010-12-30 | 车辆自适应巡航控制系统及方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012088635A1 true WO2012088635A1 (zh) | 2012-07-05 |
Family
ID=46382155
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2010/002208 WO2012088635A1 (zh) | 2010-12-30 | 2010-12-30 | 车辆自适应巡航控制系统及方法 |
Country Status (2)
Country | Link |
---|---|
US (1) | US9266533B2 (zh) |
WO (1) | WO2012088635A1 (zh) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015032508A1 (de) * | 2013-09-05 | 2015-03-12 | Avl List Gmbh | Verfahren und vorrichtung zur optimierung von fahrerassistenzsystemen |
CN109334564A (zh) * | 2018-09-11 | 2019-02-15 | 南京航空航天大学 | 一种防碰撞的汽车主动安全预警系统 |
CN109491377A (zh) * | 2017-09-11 | 2019-03-19 | 百度(美国)有限责任公司 | 用于自动驾驶车辆的基于dp和qp的决策和规划 |
CN110968088A (zh) * | 2018-09-30 | 2020-04-07 | 百度(美国)有限责任公司 | 车辆控制参数的确定方法、装置、车载控制器和无人车 |
TWI706238B (zh) * | 2018-12-18 | 2020-10-01 | 大陸商北京航跡科技有限公司 | 用於自動駕駛的系統和方法 |
WO2021204177A1 (zh) * | 2020-04-08 | 2021-10-14 | 长城汽车股份有限公司 | 一种车辆控制方法及装置 |
CN113682302A (zh) * | 2021-08-03 | 2021-11-23 | 中汽创智科技有限公司 | 一种驾驶状态估计方法、装置、电子设备及存储介质 |
CN114329306A (zh) * | 2021-12-30 | 2022-04-12 | 上海洛轲智能科技有限公司 | 制动距离的计算方法、装置、汽车及介质 |
CN114756025A (zh) * | 2022-04-02 | 2022-07-15 | 天津大学 | 一种自主小车巡航控制方法及装置 |
CN115009278A (zh) * | 2022-08-08 | 2022-09-06 | 潍柴动力股份有限公司 | 一种巡航控制方法、装置、设备及存储介质 |
US12304510B2 (en) | 2020-06-16 | 2025-05-20 | Avl List Gmbh | System for testing a driver assistance system of a vehicle |
Families Citing this family (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103718221B (zh) * | 2011-08-04 | 2016-08-17 | 丰田自动车株式会社 | 车辆用信息处理装置及车辆用信息处理方法 |
CN103963785A (zh) * | 2014-05-20 | 2014-08-06 | 武汉理工大学 | 一种用于汽车自适应巡航系统的双模式控制方法 |
CN107074958A (zh) * | 2014-07-09 | 2017-08-18 | 博笛生物科技有限公司 | 用于治疗肿瘤的抗‑pd‑l1组合 |
CN112546238A (zh) * | 2014-09-01 | 2021-03-26 | 博笛生物科技有限公司 | 用于治疗肿瘤的抗-pd-l1结合物 |
CN105109488B (zh) * | 2015-08-11 | 2017-10-20 | 奇瑞汽车股份有限公司 | 一种智能跟车系统及方法 |
US10486696B2 (en) | 2015-09-23 | 2019-11-26 | International Business Machines Corporation | Automated setting of cruising speeds |
CN105857309B (zh) * | 2016-05-25 | 2018-06-26 | 吉林大学 | 一种考虑多目标的车辆自适应巡航控制方法 |
CN107918585B (zh) * | 2016-10-07 | 2023-05-02 | 福特全球技术公司 | 用于测试软件程序的方法和装置 |
WO2019060909A1 (en) * | 2017-09-25 | 2019-03-28 | Continental Automotive Systems, Inc. | ADAPTIVE SPEED CONTROL SYSTEM AND ASSOCIATED METHOD |
US10725467B2 (en) | 2017-12-28 | 2020-07-28 | Robert Bosch Gmbh | System for personalizing the driving behavior of autonomous driving systems based on a vehicle's location |
US11107002B2 (en) * | 2018-06-11 | 2021-08-31 | Traxen Inc. | Reinforcement learning based ground vehicle control techniques |
US10816985B2 (en) * | 2018-04-17 | 2020-10-27 | Baidu Usa Llc | Method on moving obstacle representation for trajectory planning |
US10755007B2 (en) * | 2018-05-17 | 2020-08-25 | Toyota Jidosha Kabushiki Kaisha | Mixed reality simulation system for testing vehicle control system designs |
CN111045422A (zh) * | 2018-10-11 | 2020-04-21 | 顾泽苍 | 一种自动驾驶导入“机智获得”模型的控制方法 |
US10955853B2 (en) | 2018-12-18 | 2021-03-23 | Beijing Voyager Technology Co., Ltd. | Systems and methods for autonomous driving |
CN110194156B (zh) * | 2019-06-21 | 2020-11-10 | 厦门大学 | 智能网联混合动力汽车主动避撞增强学习控制系统和方法 |
JP7158356B2 (ja) * | 2019-09-25 | 2022-10-21 | 本田技研工業株式会社 | 走行制御装置 |
DE102020204082A1 (de) * | 2019-10-11 | 2021-04-15 | Robert Bosch Gesellschaft mit beschränkter Haftung | Verfahren zum Betreiben eines Fahrerassistenzsystems eines Fahrzeugs in einem Unterstützungsmodus und Fahrerassistenzsystem |
US11285949B2 (en) | 2019-12-04 | 2022-03-29 | Hyundai Motor Company | Vehicle travel control system and control method therefor |
US11352004B2 (en) | 2019-12-04 | 2022-06-07 | Hyundai Motor Company | Vehicle travel control system and control method therefor |
US11634130B2 (en) | 2020-03-26 | 2023-04-25 | Robert Bosch Gmbh | Adapting an advanced driver assistance system of a vehicle |
CN112109708B (zh) * | 2020-10-26 | 2023-07-14 | 吉林大学 | 一种考虑驾驶行为的自适应巡航控制系统及其控制方法 |
CN112406873B (zh) * | 2020-11-19 | 2022-03-08 | 东风汽车有限公司 | 纵向控制模型参数确认方法、车辆控制方法、存储介质和电子设备 |
FR3116252B1 (fr) | 2020-11-19 | 2023-03-24 | Renault Sas | Système et procédé de contrôle adapté à la perception |
CN115014798A (zh) * | 2022-05-30 | 2022-09-06 | 上海商汤临港智能科技有限公司 | 一种油门刹车标定方法、装置、计算机设备及存储介质 |
US12187279B2 (en) | 2022-09-29 | 2025-01-07 | Toyota Motor Engineering & Manufacturing North America, Inc. | Method and system for personalized car following with transformers and RNNs |
CN116494974B (zh) * | 2023-06-26 | 2023-08-25 | 北京理工大学 | 基于道路风险评估的自适应巡航控制方法、系统及设备 |
CN116620281B (zh) * | 2023-07-21 | 2023-10-20 | 科大国创合肥智能汽车科技有限公司 | 自适应巡航系统平顺性控制方法、电子设备及存储介质 |
CN120096565B (zh) * | 2025-05-12 | 2025-07-11 | 上海卫创信息科技有限公司 | 神经网络驱动的车辆自适应巡航控制方法 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20010032048A1 (en) * | 2000-04-17 | 2001-10-18 | Manfred Hellmann | Method and device for adaptive control of separation distance and/or driving speed of a motor vehicle |
CN101326074A (zh) * | 2005-12-13 | 2008-12-17 | 斯堪尼亚有限公司 | 自适应巡航控制系统 |
US20090254260A1 (en) * | 2008-04-07 | 2009-10-08 | Axel Nix | Full speed range adaptive cruise control system |
JP2009248683A (ja) * | 2008-04-03 | 2009-10-29 | Toyota Motor Corp | 車間距離制御装置 |
US20090321165A1 (en) * | 2006-04-12 | 2009-12-31 | Karsten Haug | Speed Control Device and Motor Vehicle Having Such a Speed Control Device |
WO2010116499A1 (ja) * | 2009-04-08 | 2010-10-14 | トヨタ自動車株式会社 | 車両走行制御装置 |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3470453B2 (ja) * | 1995-04-06 | 2003-11-25 | 株式会社デンソー | 車間距離制御装置 |
DE10211475A1 (de) * | 2002-03-15 | 2003-09-25 | Bosch Gmbh Robert | Verfahren zur Wahl des Betriebszustands eines Geschwindigkeitsregelsystems für Kraftfahrzeuge |
SE525479C2 (sv) * | 2003-07-10 | 2005-03-01 | Volvo Lastvagnar Ab | Metod för optimering av bromsförlopp i fordon |
JP4628683B2 (ja) * | 2004-02-13 | 2011-02-09 | 富士重工業株式会社 | 歩行者検出装置、及び、その歩行者検出装置を備えた車両用運転支援装置 |
DE102005011241A1 (de) * | 2005-03-11 | 2006-09-14 | Robert Bosch Gmbh | Verfahren und Vorrichtung zur Kollisionswarnung |
EP1997705B1 (en) * | 2005-12-28 | 2012-06-13 | National University Corporation Nagoya University | Drive behavior estimating device, drive supporting device, vehicle evaluating system, driver model making device, and drive behavior judging device |
US20080167820A1 (en) * | 2007-01-04 | 2008-07-10 | Kentaro Oguchi | System for predicting driver behavior |
JP5196006B2 (ja) * | 2009-03-09 | 2013-05-15 | トヨタ自動車株式会社 | 車両走行制御装置 |
US8744661B2 (en) * | 2009-10-21 | 2014-06-03 | Berthold K. P. Horn | Method and apparatus for reducing motor vehicle traffic flow instabilities and increasing vehicle throughput |
-
2010
- 2010-12-30 WO PCT/CN2010/002208 patent/WO2012088635A1/zh active Application Filing
- 2010-12-30 US US13/976,614 patent/US9266533B2/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20010032048A1 (en) * | 2000-04-17 | 2001-10-18 | Manfred Hellmann | Method and device for adaptive control of separation distance and/or driving speed of a motor vehicle |
CN101326074A (zh) * | 2005-12-13 | 2008-12-17 | 斯堪尼亚有限公司 | 自适应巡航控制系统 |
US20090321165A1 (en) * | 2006-04-12 | 2009-12-31 | Karsten Haug | Speed Control Device and Motor Vehicle Having Such a Speed Control Device |
JP2009248683A (ja) * | 2008-04-03 | 2009-10-29 | Toyota Motor Corp | 車間距離制御装置 |
US20090254260A1 (en) * | 2008-04-07 | 2009-10-08 | Axel Nix | Full speed range adaptive cruise control system |
WO2010116499A1 (ja) * | 2009-04-08 | 2010-10-14 | トヨタ自動車株式会社 | 車両走行制御装置 |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10399565B2 (en) | 2013-09-05 | 2019-09-03 | Avl List Gmbh | Method and device for optimizing driver assistance systems |
CN105579320A (zh) * | 2013-09-05 | 2016-05-11 | 李斯特内燃机及测试设备公司 | 一种用于优化驾驶员辅助系统的方法和设备 |
WO2015032508A1 (de) * | 2013-09-05 | 2015-03-12 | Avl List Gmbh | Verfahren und vorrichtung zur optimierung von fahrerassistenzsystemen |
CN105579320B (zh) * | 2013-09-05 | 2020-08-14 | 李斯特内燃机及测试设备公司 | 一种用于优化驾驶员辅助系统的方法和设备 |
CN109491377B (zh) * | 2017-09-11 | 2022-07-08 | 百度(美国)有限责任公司 | 用于自动驾驶车辆的基于dp和qp的决策和规划 |
CN109491377A (zh) * | 2017-09-11 | 2019-03-19 | 百度(美国)有限责任公司 | 用于自动驾驶车辆的基于dp和qp的决策和规划 |
CN109334564B (zh) * | 2018-09-11 | 2021-04-02 | 南京航空航天大学 | 一种防碰撞的汽车主动安全预警系统 |
CN109334564A (zh) * | 2018-09-11 | 2019-02-15 | 南京航空航天大学 | 一种防碰撞的汽车主动安全预警系统 |
CN110968088B (zh) * | 2018-09-30 | 2023-09-12 | 百度(美国)有限责任公司 | 车辆控制参数的确定方法、装置、车载控制器和无人车 |
CN110968088A (zh) * | 2018-09-30 | 2020-04-07 | 百度(美国)有限责任公司 | 车辆控制参数的确定方法、装置、车载控制器和无人车 |
TWI706238B (zh) * | 2018-12-18 | 2020-10-01 | 大陸商北京航跡科技有限公司 | 用於自動駕駛的系統和方法 |
WO2021204177A1 (zh) * | 2020-04-08 | 2021-10-14 | 长城汽车股份有限公司 | 一种车辆控制方法及装置 |
US12304510B2 (en) | 2020-06-16 | 2025-05-20 | Avl List Gmbh | System for testing a driver assistance system of a vehicle |
CN113682302A (zh) * | 2021-08-03 | 2021-11-23 | 中汽创智科技有限公司 | 一种驾驶状态估计方法、装置、电子设备及存储介质 |
CN114329306A (zh) * | 2021-12-30 | 2022-04-12 | 上海洛轲智能科技有限公司 | 制动距离的计算方法、装置、汽车及介质 |
CN114756025A (zh) * | 2022-04-02 | 2022-07-15 | 天津大学 | 一种自主小车巡航控制方法及装置 |
CN115009278B (zh) * | 2022-08-08 | 2022-11-29 | 潍柴动力股份有限公司 | 一种巡航控制方法、装置、设备及存储介质 |
CN115009278A (zh) * | 2022-08-08 | 2022-09-06 | 潍柴动力股份有限公司 | 一种巡航控制方法、装置、设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
US20140012479A1 (en) | 2014-01-09 |
US9266533B2 (en) | 2016-02-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2012088635A1 (zh) | 车辆自适应巡航控制系统及方法 | |
CN102109821B (zh) | 车辆自适应巡航控制系统及方法 | |
CN112677983B (zh) | 一种识别驾驶员驾驶风格的系统 | |
CN107808027B (zh) | 基于改进模型预测控制的自适应跟车方法 | |
CN108437991B (zh) | 一种智能电动汽车自适应巡航控制系统及其方法 | |
Schnelle et al. | A driver steering model with personalized desired path generation | |
JP6970117B2 (ja) | 運転者のルールベース型支援のための制御データ作成方法 | |
CN103085816B (zh) | 一种用于无人驾驶车辆的轨迹跟踪控制方法及控制装置 | |
CN104834776B (zh) | 一种微观交通仿真中交通车辆建模仿真系统及方法 | |
CN111775949A (zh) | 一种人机共驾控制系统的个性化驾驶员转向行为辅助方法 | |
Marcano et al. | Low speed longitudinal control algorithms for automated vehicles in simulation and real platforms | |
EP2006177A2 (en) | Vehicle speed control apparatus in accordance with curvature of vehicle trajectory | |
WO2014156256A1 (ja) | 車両の運動制御装置 | |
CN111439264B (zh) | 一种基于人机混驾的换道控制模型的实现方法 | |
Zhao et al. | Model-free optimal control based intelligent cruise control with hardware-in-the-loop demonstration [research frontier] | |
JP2005178627A (ja) | 車両の統合制御システム | |
CN103764471A (zh) | 车辆控制装置 | |
CN111873975B (zh) | 一种电子驻车制动的控制方法、装置、系统、设备及介质 | |
CN110103960B (zh) | 车辆自适应巡航控制方法、系统及车辆 | |
CN112109705A (zh) | 增程式分布驱动电动车辆避撞优化控制系统及控制方法 | |
CN108515971A (zh) | 一种巡航功能控制方法、系统、装置及可读存储介质 | |
CN114771520B (zh) | 一种基于强化学习的电动汽车经济性自适应巡航控制方法及系统 | |
CN115123159A (zh) | 一种基于ddpg深度强化学习的aeb控制方法及系统 | |
CN101264762A (zh) | 车辆跟驰运动的速度控制方法 | |
GB2498429A (en) | In-vehicle training method/system for teaching fuel economy |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10861291 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13976614 Country of ref document: US |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 10861291 Country of ref document: EP Kind code of ref document: A1 |