61 results about "Dual carriageway" patented technology

The invention discloses an automatic driving overtaking decision-making method based on reinforcement learning under opposite double lanes. The method comprises the following steps of collecting the traffic state of an automatic driving vehicle through a sensor; inputting the collected traffic state into a trained decision model; enabling the decision model to select a corresponding driving actioninstruction from the action space according to the input information and output the driving action instruction, and driving the vehicle automatically to form a new traffic state after the driving action; calculating a reward value of the driving action through a reward function, and storing the original traffic state, the driving action, the reward value and the new traffic state into an experience playback pool as transfer samples; calculating a loss function value of the decision model, and optimizing parameters of the decision model according to the transfer sample and the loss function value; and repeating the steps until the automatic driving is finished. The safety and comfort of the overtaking decision-making process of the automatic driving vehicle are ensured, and the humanization and robustness of decision-making are improved through a reinforcement learning decision-making method.

The invention relates to a double-lane intelligent network connection vehicle cooperative lane changing method, device and equipment and a storage medium. The method comprises the steps that motion state information of vehicles around a lane changing vehicle is obtained in real time, and a transverse acceleration model and a longitudinal acceleration model of the lane changing vehicle M are constructed. According to the models, multiple constraints such as safety, comfort and a state target of lane changing are considered, and the target of safe cooperative lane changing of the intelligent network connection vehicle is achieved. The whole lane changing process is divided into two stages, the purpose of actively adjusting the traffic flow of the two lanes is achieved by actively controllinga target lane and a main lane to cooperate with the lane changing vehicle in the lane changing target stage, the negative influence of lane changing on the traffic flow of the two lanes is improved to a certain extent, and the road passing efficiency of the two lanes is improved; and in consideration of the fact that a feasible solution does not exist under certain lane changing initial conditions when fixed lane changing time is adopted, the collaborative lane changing time is selected as a range value, the feasibility of optimizing model solving is effectively improved, and then the feasible lane changing range is widened.

The invention relates to a mountainous area double-lane highway hidden danger road section traffic conflict risk dynamic evaluation method, which comprises the following steps: firstly, obtaining vehicle driving track data and traffic flow data of a current implementation road section, calculating according to the track data to obtain a serious conflict discrimination threshold, and extracting thetraffic flow data 5 minutes before the occurrence moment of a serious traffic conflict; selecting mountainous area double-lane road traffic conflict risk dynamic possibility evaluation indexes and conflict severity evaluation indexes, establishing a road conflict risk evaluation model, and determining conflict risk possibility and conflict severity grades; and combining conflict risk possibilitygrading and conflict severity grading to construct a mountainous area double-lane highway traffic conflict risk dynamic evaluation comprehensive matrix, and determining the conflict risk grade of eachtraffic conflict of the current implementation road section. According to the method, the collected high-precision space-time trajectory data and traffic flow data are used for dynamically evaluatingthe traffic conflict risk of the mountainous area double-lane road, the data are easy to obtain, and the constructed model has relatively good prediction precision.

The invention relates to the field of V2V equipment testing, in particular to a V2V-based test method and test system for a vehicle emergency braking early warning system. The test method comprises the following steps: constructing a test case which at least comprises test road conditions of a straight lane, a curve and a ramp of a single lane and test road conditions of a straight lane and a curve of two adjacent lanes; driving the target vehicle and the background vehicle on the test road condition according to the test case, recording early warning information sent by the target vehicle, and recording running time information of the target vehicle and the background vehicle; and analyzing whether the early warning information meets a test index or not according to the runtime information. The test system simulates a test scene through the test case construction module, and is wide in coverage and high in pertinence; the running time information of the target vehicle and the background vehicle is obtained through the test module, the synchronism is high, and the accuracy is high; and a test index is generated through the first test module. On the whole, according to the test method and system, the simulated test scene is complete, the key performance evaluation indexes of the system are mature, and the test reliability is high.

The invention discloses a bridge dynamic load automatic detection and inversion system and method. The system comprises a three-dimensional laser scanner (11), a computer, a plurality of cameras (12)and a timer used in cooperation with the three-dimensional laser scanner (11). A plurality of road rods are uniformly distributed on a same-direction lane on the bridge; the cameras (12) for coveringthe view field of a lane area between two adjacent road rods are erected on all road rods, the plurality of cameras are used for synchronously shooting, the synchronization precision is less than 10 microseconds, the three-dimensional laser scanner (11) is erected right above two lanes in the same direction at the entrance of a bridge, and the measurement range covers the two lanes; a stress sensor is at least arranged at the maximum stress position of the bridge (21), and the stress sensor, the cameras (12), the timer and the three-dimensional laser scanner (11) are all connected with a computer. According to the method, dynamic load detection of the bridge and accurate monitoring of strain of each part of the bridge are realized.

The invention discloses a double-lane type battery replacement station, which comprises two groups of guiding positioning lanes arranged oppositely and at an interval and a charging chamber arranged between the two groups of guiding positioning lanes. The charging chamber comprises: a first charging rack; and a second charging rack, which is opposite to the first charging rack and at an interval in order to form a conveying channel between the first charging rack and the second charging rack. A stacker crane for taking and placing batteries is disposed in the conveying lane, and at least one battery conveying trolley, which moves back and forth between the stacker crane and the guiding positioning lane, is arranged between the stacker crane and the guiding positioning lane. According to the invention, the double-lane layout is adopted, the charging and storage capacity of the single group of charging chamber can be fully utilized, the battery replacement efficiency is improved, and theequipment cost and the land use cost are reduced; the movement stroke of the charging connector is improved to improve the butt joint efficiency, and meanwhile, the capacity of the charging chamber for accommodating the batteries is greatly improved; the automation degree of the battery replacement process is improved, and the battery replacement efficiency is further improved.

The invention relates to the technical field of intelligent traffic, in particular to an ETC channel intelligent passing system and method. The system comprises an outlet double-lane isolation systemand an automatic card taking machine. The outlet double-lane isolation system comprises a left isolation fence, a right isolation fence, a second barrier gate, a third barrier gate and a fourth barrier gate. A double-lane area is jointly isolated by the left isolation fence, the right isolation fence, the second barrier gate and the fourth barrier gate in front of an exit of an ETC channel, the third barrier gate divides the double-lane area into a left side straight lane and a right side emergency lane, and the automatic card taking machine is arranged in the right side emergency lane. When the situation that an ETC user cannot conduct normal transaction at the ETC channel happens, a background management terminal controls the first barrier gate and the third barrier gate to release, a vehicle can enter the ETC channel without parking, the right side emergency lane is utilized for card taking or swiping for passing, traffic jam and safety accidents are effectively avoided, the passingexperience of the ETC user is improved, and rapid and smooth passing of the ETC channel is guaranteed.

The invention discloses a mountainous-area two-lane curve active early-warning anti-collision guard bar and a rear-end collision early-warning method. Through the anti-collision guard bar, on the condition that the linear change of a mountainous-area curve is dramatic and the visibility range is poor, vehicles in two two-way lanes can be monitored in real time, the dynamic position change of all the vehicles is monitored, and then dangerous conditions which possibly occur can be subjected to early warning. The method comprises the steps of prompting of the running conditions of the vehicles ofopposite-direction lanes, early warning of rear-end collision of the front and rear vehicles, and prompting early warning through an acousto-optic system. By means of the mountainous-area two-lane curve active early-warning anti-collision guard bar and the rear-end collision early-warning method, main traffic risks which possibly occur in the curve can be subjected to all-directional and systematic advance early warning, and thus the occurrence of accidents is reduced; meanwhile, active early warning and passive anti-collision facility integrated design is conducted, and double guarantees ofthe traffic safety are achieved.

The invention provides a device for cleaning ice and snow on a road surface through laser light. The device comprises a protection unit (1), a laser scanning unit (2), a storage battery (3), a transport vehicle (4) and a liquefied water treating unit (5), wherein the storage battery (3) is connected with the protection unit (1) and the laser scanning unit (2), and is used for supplying power to the protection unit (1) and the laser scanning unit (2); the laser scanning unit (2) is used for outputting a linear focused ray; and the ice and the snow on the road surface are cleaned through the movement of the transport vehicle (4). The device has the advantages of high ice and snow removing efficiency, low cost, good effect, no pollution and the like. The requirement of removal of ice and snow from a sidewalk can be met by carrying through a manual trolley; and the lengths of scanning lines can be prolonged by splicing a plurality of laser light scanning units, so that the requirement of parallel ice and snow removing task on a single lane or dual lanes is met.

The invention discloses a method for improving Internet of Vehicles communication connectivity based on double-lane grid fluid mechanics. The method comprises the following steps: 1) constructing a double-lane Internet of Vehicles communication model; 2) deducing a density condition which needs to be met for ensuring the stability of the traffic flow; 3) constructing a direct connectivity probability function based on traffic flow stability; 4) deducing conditions for stabilizing the traffic flow after the driver adjusts the time in the traffic flow changing process; 5) constructing a continuous connectivity probability function between the vehicles. According to the method, the movement of the vehicles on the road is regarded as the flow of particles, and the lane changing behaviors of the vehicles are coordinated by adopting a double-lane grid fluid mechanics model, so that traffic jam caused by a large number of lane changing behaviors is avoided, and the communication connectivity of the vehicles is improved on the basis of ensuring the stability of the traffic flow.

The invention discloses a vehicle automatic lane changing control method, which comprises the following steps of S1, constructing a double-lane five-vehicle road scene model, and defining pose coordinates and a state space of a vehicle; and S2, constructing a control strategy from three aspects of traceability, safety and comfort, and completing a trajectory tracking task, wherein the step comprises four sub-steps of traceability performance design, safety performance design, comfort performance design and controller design. According to the vehicle automatic lane changing control method, theautomatic lane changing capacity in the road environment can be provided for the vehicle, the lane changing controller is designed through model prediction control, the road scene model is constructed, and coordination of the vehicle tracking performance, the safety performance and the comfort performance is achieved.

The invention discloses an expressway double-lane cooperative control method in a mixed traffic scene, which comprises two lanes of an expressway in the same driving direction, and the two lanes comprise a first lane and a second lane; a plurality of HVs (human driving vehicles) and a plurality of CAVs (networked automatic vehicles) run on the two lanes, and the method is characterized by comprising the following steps: S1, defining an expressway mixed traffic sub-motorcade; S2, performing communication topology of the networked automatic vehicles in the sub-vehicle teams; S3, constructing a networked automatic vehicle control strategy and a human driving model; and S4, adopting a two-lane mixed traffic group control method based on spatial distribution. The invention provides a method for dividing mixed traffic sub-motorcades according to the physical space positions of the networked automatic vehicles and the human driving vehicles on the road, and constructs a new networked automatic vehicle control algorithm based on the method, so that the control algorithm of the networked automatic vehicles in a mixed traffic scene is easier to implement.

The invention discloses a double-lane vehicle lane changing guiding method under vehicle-road cooperation, and the method comprises the steps: describing the state information of a vehicle based on acellular automaton model, carrying out the real-time effective evaluation of the threat to the vehicle in real time according to the information of surrounding vehicles, carrying out the traversal ofthe elements of a threat evaluation function, obtaining the position of a minimum value, and giving the vehicle lane changing guidance; in a double-lane environment, subjecting the observation vehicleto vehicle speed induction according to the movement trend of the forward vehicle after lane changing, so that the vehicle speed of the observation vehicle is associated with the acceleration trend of the forward vehicle after lane changing; referring to the motion trend of the backward vehicle, observing that the vehicle adopts a leading acceleration motion rule, and avoiding rear-end collisionaccidents while ensuring motion synchronization; providing early-stage speed guidance for the vehicle close to the blocking point, and avoiding the phenomenon of vehicle starting and stopping; and finally, giving a double-lane vehicle lane changing rule under vehicle-road cooperation to complete lane changing. According to the invention, the problems of vehicle congestion and road section congestion point increase in the future traffic environment are solved.

The invention relates to an autonomous driving method and system. The following characteristics of the automatic driving vehicle are analyzed; constructing a following model; the lane changing characteristics of the automatic driving vehicle are analyzed; lane changing model simulation and artificial driving vehicle lane changing effects are achieved; in combination with influence factors of lanechanging of the automatic driving vehicle, lane changing characteristics and characteristics of an autonomous driving simulation platform, speed bearing degree and space allowance degree are introduced to constrain lane changing behaviors so as to construct a matching model, lane changing space is confirmed, and a lane changing preparation model is constructed; after the lane changing behavior isgenerated, according to the actual space size of the current lane change and considering the actual situation of information processing time delay, a lane changing speed control model based on the double-lane front driving vehicle and a track optimization model based on lane changing space are constructed, and it is ensured that the operation parameters generated through the autonomous driving model are accurately attached to the traffic scene where the automatic driving vehicle is actually located.

The invention discloses a tool for lifting big pieces for double travelling cranes. The tool comprises a main beam, wherein the main beam is formed in a way that two parallel H-shaped beams are weldedalong upper and lower flanges, and a plurality of lining plates are welded in the U-shaped position of the flange, which is welded, on each H-shaped beam; two travelling crane lifting lugs are symmetrically welded above the main beam from left to right along the longitudinal center line of the main beam, and a big piece lifting lug is welded in the middle below the main beam; and rib plates are symmetrically welded respectively on both sides of the two travelling crane lifting lugs and the big piece lifting lug, and the bottoms of the rib plates are fixedly welded with the upper surface of the flanges of the H-shaped beams of the main beam. The invention further discloses a manufacturing method of the tool for lifting the big pieces for the double travelling cranes. By applying the tool,two travelling cranes can be used for lifting the big pieces in parallel, the tool has the advantages of rapidity, convenience, safety, low investment cost and high working efficiency, the onsite actual demand can be satisfied, and the production cost is reduced.

The invention relates to a lane obstacle avoidance method and device for an automatic driving vehicle, and belongs to the technical field of intelligent vehicle driving. According to the state of the obstacle and the distance between the vehicle and the obstacle, setting a path planning reference point; substituting the path planning reference point into a set Bezier curve equation, and planning a reference driving path based on a Bezier curve; and controlling the vehicle to drive according to the reference driving path to realize obstacle avoidance driving. According to the lane obstacle avoidance method and device, when the vehicle encounters the obstacle, the vehicle can run along the current lane in an obstacle avoidance mode, unnecessary obstacle avoidance and lane changing behaviors are avoided as much as possible, and the real-time performance, rationality and safety of behavior decision making are improved; under the double-lane working condition, obstacle avoidance returning is preferentially selected, and when obstacle avoidance returning of the lane cannot be achieved, an obstacle avoidance lane changing strategy is provided, so that the obstacle avoidance flexibility is improved.

The invention discloses a road full-traversal sweeping path planning method of an unmanned sweeper. The method comprises the following steps that the unmanned sweeper judges the type of a to-be-sweptroad; if the lane is a single lane, autonomously identifying the road edge, the road edge is marked as a welt cleaning area, the welt cleaning area is preferentially cleaned according to related traffic rules, and a spiral inward cleaning path is planned; if the vehicle is in a double-lane or multi-lane mode, the road edge and the middle lane line or the middle isolation belt are autonomously recognized, and a sweeping path is planned according to the principle that the welt sweeping area is swept preferentially, the middle lane line area or the middle isolation belt area is swept secondly andthe area in the lane is swept finally. Planning of various paths complies with traffic rules, no retrograde driving risk exists, the vehicle cannot press or cross a middle lane line or a middle isolation belt to drive, and driving safety is guaranteed; the spiral inward path and the sweeping sequence can guarantee driving safety, sweeping missing and repeated sweeping are avoided, and the sweeping efficiency is improved.

The invention discloses a following vehicle state change estimation method considering transverse interference under double-lane driving. The following vehicle state change estimation method comprises the steps that S01, a simulation analysis environment is constructed under the double-lane driving condition; s02, acquiring vehicle track data and lane geometric dimension data, inputting the data into a simulation analysis environment, and performing vector decomposition to obtain a cut-in vehicle speed vector and a distance vector decomposition result; s03, in the simulation analysis environment, when the cut-in vehicle starts to enter the target lane, state changes between the cut-in vehicle and the following vehicles are estimated; s04, calculating an interference factor and an attenuation factor in a simulation analysis environment; s05, obtaining the influence degree of the transverse interference on the following vehicle and the correlation characteristics between the front vehicle and the rear vehicle in the following vehicle team; and S06, estimating the residual area of the lane according to the distance vector. According to the method, the transverse interference to the following vehicle when the vehicle cuts into the lane is considered, and the precision of state change analysis of the following vehicle can be effectively improved.

The invention discloses a vehicle motion simulation analysis method in an urban traffic scene. The intelligent following model based on the safety distance is improved, the characteristics that the vehicle safety distance is small and the following vehicle starts to follow the front vehicle to accelerate in an urban traffic scene can be better simulated by adding an acceleration adjustment item toa vehicle acceleration formula, and the applicability of the model is improved. A double-lane model based on acceleration analysis is improved, the lane changing process of a vehicle is evaluated tojudge whether lane changing behaviors can be executed or not, and the rationality of the proposed model is verified through experiments.

The invention discloses a plateau area double-lane road curve slope combination section alignment index selection method. The method comprises the following specific steps that collecting vehicle speed, heart rate and electroencephalogram signal data of a driver in different curve slope combination sections in combination with an on-site driving test; performing sample entropy processing on the electroencephalogram and heart rate indexes of the driver, and extracting feature indexes of the electroencephalogram and heart rate indexes; performing principal component analysis on the heart rate sample entropy and the electroencephalogram signal sample entropy to obtain a driver psychological load comprehensive evaluation index CISE; constructing a regression model of the comprehensive evaluation index CISE and the linear combination values CA of the curve slope combination sections of the different altitude intervals; and finally, determining linear combination thresholds Cacri corresponding to different altitude intervals by utilizing the Lagrange median theorem, and taking the linear combination thresholds Cacri as linear design indexes of the curve slope combination road section. According to the method, the linear design index selection problem of the curve slope combined road section under the plateau low-pressure oxygen-deficient environment is solved, and theoretical support is provided for linear safety design of the plateau road.

The invention discloses a method for monitoring the running state of a tunnel based on a deep neural network, and the method comprises the steps: detecting a vehicle in a monitoring image, which is smaller than a foremost vehicle threshold value, and enabling the vehicle not to complete a complete overtaking operation in the range of the foremost vehicle threshold value, according to the method and the system, the condition that the foremost vehicle changes but the vehicle does not pass due to overtaking is avoided, the foremost vehicle in the two lanes is obtained and only the foremost vehicle is judged, data calculation is reduced, meanwhile, only the foremost clear vehicle is recognized during color recognition and license plate recognition, and the accuracy of vehicle recognition is greatly improved.

The invention discloses a 5G-based mining area unmanned transportation system marshalling cooperative operation method which comprises the steps: setting a roadside unit at a meeting position, installing a vehicle-mounted unit and a 5G-based multi-vehicle cooperative positioning system for obtaining the position, speed and acceleration information of vehicles on each vehicle, and a central controlplatform; judging the time when two vehicles move to the meeting position of the two lanes according to the current vehicle speed by calculating, and controlling and adjusting the vehicle speed within an allowable range under the large principle of full-running and empty-parking to complete cooperative marshalling transportation of upper and lower mine trucks. According to the technical scheme, acollaborative marshalling scheme is provided for geologic one-way driverless mineral transportation of a mining area, and based on the characteristics of high speed and low time delay of 5G, information such as the position and the speed of a mining truck can be received and calculated immediately in time; and meanwhile, the unmanned transportation efficiency of the mining area is directly improved by sending and executing demand control.

The invention belongs to the technical field of curb construction, and particularly relates to a one-time pouring construction process for a ramp double-lane separation curb and a main body structure.The process comprises the following steps of S1, binding reinforcing steel bars of an inclined plate main body structure according to a traditional construction process; S2, when a fourth joint is constructed, installing embedded parts, and when a fifth joint is constructed, binding the reinforcing steel bars; S3, erecting a ramp separation curb template; S4, mounting a separation curb concrete chamfering mold; S5, carrying out concrete pouring on a ramp inclined plate main body structure, and carrying out ramp separation curb concrete pouring immediately after pouring; and S6, when the fourth joint is constructed, installing a finished shading plate, and when the fifth joint is constructed, smearing cement mortar for leveling and troweling. The method solves problems of a long construction period and difficulty in construction of the existing curb, and can be widely applied to construction of various pavements.

The invention discloses a method for determining the minimum safe radius of a horizontal curve suitable for a double-lane road in a plateau area. The method comprises the following steps: (1) acquiring vehicle speed, heart rate and electroencephalogram data of a driver driving on horizontal curve road sections with different altitudes and radiuses according to an on-site driving experiment; (2) performing sample entropy processing on the electroencephalogram and heart rate indexes of the driver, and extracting feature indexes of the electroencephalogram and heart rate indexes; (3) performing principal component analysis on the heart rate sample entropy and the electroencephalogram signal sample entropy to obtain a driver psychological load comprehensive evaluation index CISE; (4) constructing a regression model of the comprehensive evaluation index CISE and the horizontal force acceleration of the horizontal curve road sections in different altitude intervals; and (5) determining the threshold value of the comprehensive evaluation index through the Lagrange median theorem, and finally proposing the minimum safety radius threshold value of the double-lane road in different altitude intervals in the plateau region. The driving safety of highways in the plateau area is improved.

The present invention relates to a method, device, equipment and storage medium for coordinated lane change of two-lane intelligent networked vehicles. The method obtains the motion state information of vehicles around the lane-changing vehicle in real time, and constructs a lateral acceleration model of the lane-changing vehicle M. And the longitudinal acceleration model, which considers multiple constraints such as safety, comfort and state goals of lane changing, and realizes the goal of safe and coordinated lane changing of intelligent networked vehicles. The present invention divides the whole lane changing process into two stages, and in the lane changing target stage, the purpose of actively adjusting the traffic flow of the two lanes is achieved by actively controlling the target lane and the main lane to cooperate with the lane changing vehicles, and the impact of lane changing on the traffic flow of the two lanes is improved to a certain extent. to improve the traffic efficiency of two-lane roads; considering that the fixed lane change time may not have a feasible solution under certain lane change initial conditions, the cooperative lane change time is selected as the range value to effectively improve the feasibility of the optimization model solution. Thereby increasing the feasible range of lane changing.