CN111708372B - Self-adaptive safe driving method and system for bumpy road surface of unmanned sweeper - Google Patents
Self-adaptive safe driving method and system for bumpy road surface of unmanned sweeper Download PDFInfo
- Publication number
- CN111708372B CN111708372B CN202010855733.9A CN202010855733A CN111708372B CN 111708372 B CN111708372 B CN 111708372B CN 202010855733 A CN202010855733 A CN 202010855733A CN 111708372 B CN111708372 B CN 111708372B
- Authority
- CN
- China
- Prior art keywords
- bumpy
- sweeper
- road section
- bump
- grade
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 241001417527 Pempheridae Species 0.000 title claims abstract description 164
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000000694 effects Effects 0.000 claims abstract description 20
- 238000010408 sweeping Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims description 49
- 238000004140 cleaning Methods 0.000 claims description 31
- 238000012549 training Methods 0.000 claims description 21
- 238000012937 correction Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 7
- 239000004575 stone Substances 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 238000012790 confirmation Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 230000003245 working effect Effects 0.000 abstract 1
- 239000013598 vector Substances 0.000 description 4
- 239000011449 brick Substances 0.000 description 3
- 239000004568 cement Substances 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000004148 unit process Methods 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Electromagnetism (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a self-adaptive safe driving method and a self-adaptive safe driving system for bumpy road surfaces of an unmanned sweeper, wherein a monocular camera, a binocular camera, an ultrasonic sensor, an IMU (inertial navigation Unit) and three vibration sensors are fused to identify the bumpy road sections, the bumpy road sections are graded according to the traveling limit, the vibration condition and the sweeping effect of the sweeper, and the sweeper is controlled to make driving strategies of avoiding, decelerating and passing or directly passing aiming at the bumpy road surfaces with different bump grades, so that the sweeper can self-adaptively adjust the driving strategies according to the bump grade, and the sweeping coverage and the working effect of the sweeper are ensured to the maximum limit while the sweeper safely passes through the bumpy road sections.
Description
Technical Field
The invention relates to the technical field of sweeper vehicles, in particular to a self-adaptive safe driving method and a self-adaptive safe driving system for bumpy road surfaces of an unmanned sweeper vehicle.
Background
As a brand new generation of intelligent unmanned environmental sanitation products, unmanned cleaning vehicles are gradually favored and applied in the environmental sanitation industry due to the advantages of autonomous driving, autonomous timing cleaning task execution, high operation efficiency, long operation time and the like. But the unmanned cleaning vehicle is limited by factors such as immature unmanned technology, poor environmental adaptability and the like, and the scenes of the unmanned cleaning vehicle in domestic application are very limited. For example, the common application environments of unmanned sweeper vehicles include urban roads, parks, parking lots, parks and other places, and the road conditions of the above places often have bad road conditions such as crushed stones, deceleration strips, seRegularia Regulacea, potholes and the like, so that the sweeper vehicle cannot make correct judgment and is trapped in a dangerous condition when experiencing such road conditions, and if avoidance strategies are carried out on all recognized bumpy roads, the sweeping effect of the sweeper vehicle is greatly influenced, and even the sweeper vehicle cannot run.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a self-adaptive safe driving method and a self-adaptive safe driving system for a bumpy road surface of an unmanned sweeper, which can control the sweeper to pass through bumpy road sections with different bumping grades according to different driving strategies, ensure that the sweeper passes through the bumpy road sections and simultaneously ensure the sweeping coverage and the operation effect of the sweeper to the maximum extent.
The invention is realized by the following technical scheme: a self-adaptive safe driving method for bumpy road surfaces of an unmanned sweeper is characterized by comprising the following steps: the method comprises the following steps:
s1, acquiring an image of a road condition at a position 10-15 meters away from the advancing direction of the sweeper by a monocular camera, identifying and processing the image of the road condition by a monocular identification unit to identify the jolting type and the jolting section size of the jolting road section, obtaining the influence degree of the jolting road section on the operation of the sweeper according to the jolting type and the jolting section size, and sending the influence degree of the jolting type, the jolting section size and the jolting road section to a main control module;
s2, the main control module judges whether the influence degree of the bumpy road section on the operation of the sweeper reaches a preset safety passing threshold, if so, a deceleration instruction is sent to the control module to control the sweeper to decelerate to move forwards, if not, an original speed driving instruction is sent to the control module to control the sweeper to continue to move forwards at the original speed, and the information of the bump type and the size of the bumpy road section is reported to the cloud background system;
s3, when the sweeper moves forwards to a distance of 4-5 m from the bumpy road section, the images of the bumpy road section are collected by the binocular camera, the images of the bumpy road section are identified and processed by the binocular identification unit to obtain the bump type, the size and the depth of the bumpy road section, and the bump type, the size and the depth of the bumpy road section are sent to the main control module;
s4, the main control module judges and verifies whether the bump type obtained by the recognition processing of the monocular camera and the binocular camera is consistent, if so, the bump type recognition of the monocular camera to a bump road section is accurate, the bump type recognition confidence coefficient of the monocular camera is improved, if not, the bump type obtained by the recognition of the binocular camera is taken as the standard, the recognition information is stored in the monocular camera for the learning and correction of the monocular camera, the bump type recognition accuracy of the monocular camera is improved, in addition, the bump road section size obtained by the processing of the monocular camera is compared with the bump road section size obtained by the processing of the binocular camera, whether the difference between the two is in the size threshold range is judged, if so, the recognition of the monocular camera is judged to be accurate, the confidence coefficient of the monocular camera is improved, if not, the bump road section size obtained by the recognition of the binocular camera is taken as the standard, and the bump road section size is fed back to the monocular camera, the monocular camera is used for correction and learning, and the recognition accuracy of the monocular camera is improved;
s5, the main control module identifies the bump grade of the bump road section according to the bump road section depth, the bump type after judgment and verification and the bump road section size;
and S6, the main control module sends the driving strategy corresponding to the bumping grade to the control module according to the bumping grade identified in the step S5, and the control module controls the sweeper to pass through the bumping road section according to the corresponding driving strategy.
Further: the classification of the bump types is as follows:
class a jounce: bump-like road sections exist permanently like bump road sections such as a convex speed bump, a sewer well cover, a road section made of specific pavement materials, a road section made of specific terrains and the like;
class B jounce: a repairable bumpy road section typified by a defective-type pothole on the road surface;
class C jounce: temporary bumpy road sections such as a broken stone road section and a residual branch road section;
further: the bump rating is classified as follows:
the first-level bump refers to a bumpy road condition seriously endangering the running safety of the sweeper. Bumping which the sweeper cannot pass through and bumping which is easy to cause the damage of key parts of the sweeper or the vibration of the whole sweeper;
secondary bumping: the road condition is bumpy, which may endanger the operation safety of the sweeper and seriously affect the sweeping effect of the sweeper, the sweeper must slowly pass through the road condition at a speed lower than a limited speed to ensure the bumping of the walking safety, and the sweeping effect of the sweeper is extremely poor;
third-order jounce: the cleaning device is a bumping condition which has no obvious influence on the operation safety of the sweeper even if the cleaning device lasts for a period of time (such as 20S), but has a certain influence on the cleaning effect;
further: the driving strategy described in step S6 is as follows:
when the identified bumping grade is first-level bumping, the sweeper starts an obstacle avoidance strategy, local path planning for bypassing a bumpy road section is carried out on the basis of original path planning, after obstacle avoidance is completed, the original planning is returned to continue executing tasks, information is reported to a cloud background system, information of bumpy road conditions and avoidance results are reported: successfully and actually shooting 5 images; if the road condition that the obstacle cannot be avoided is met, the sweeper stops, backs up, turns around and returns to the storehouse, information is reported to the cloud background system, and the reported information comprises jolting road condition information and avoidance results: 5 failed images are photographed;
when the identified bumping grade is two-stage bumping, the sweeper starts an obstacle avoidance strategy, local path planning for bypassing a bumpy road section is carried out on the basis of original path planning, after obstacle avoidance is completed, the original planning is returned again to continue executing tasks, information is reported to a cloud background system, and the reported information comprises bumpy road condition information and a passing mode: avoiding, bypassing and actually shooting 5 images; if the road condition that the obstacle cannot be avoided is met, the sweeper stops the cleaning function, decelerates to a limited speed, and reports information to the cloud background system, wherein the reported information comprises bumpy road condition information and a passing mode: 5 images are passed through and photographed at a reduced speed;
when the recognized bumping grade is three-level bumping, the sweeper decelerates to a limited speed, slowly passes through a bumping road section, the coverage rate of the cleaning operation is guaranteed to the maximum extent, the cleaning operation state is kept, information is reported to a cloud background system, and the reported information comprises bumping road condition information and a passing mode: 5 pictures are taken in real time after passing through the speed reduction;
the bumpy road condition information comprises the bumpy grade, the bumpy type and the GPS position of the bumpy road section.
Further: when the sweeper slows down and passes through a bumpy road section with two-level bump and three-level bump and runs 40-60cm before the bumpy road section, the method further comprises the following steps:
(1): detecting the depth information of the bumpy road section through an ultrasonic sensor, and sending the depth information to a depth recognition unit for processing to obtain the depth of the bumpy road section;
(2): comparing the depth of the bumpy road section detected by the ultrasonic sensor with the depth of the bumpy road section processed by the binocular camera, judging whether the difference between the two is within a depth threshold value, if so, judging that the binocular camera is accurately identified, improving the confidence coefficient of the binocular camera, if not, regarding the information of the depth of the bumpy road section detected by the ultrasonic sensor as the standard, re-identifying the grade of the bumpy road section according to the bumpy road section depth detected by the ultrasonic sensor and the size of the bumpy road section after verification in step S4 to obtain the latest bumpy grade, judging whether the latest bumpy grade meets or is lower than the bumpy grade identified in step S5, if so, continuing to move the sweeper according to the driving strategy corresponding to the bumpy grade identified in step S5, and if not, emergently decelerating the sweeper according to the driving strategy corresponding to the latest bumpy grade, and the depth information of the bumpy road section detected by the ultrasonic sensor is fed back to the binocular camera for correction and learning of the binocular camera, so that the identification accuracy of the binocular camera is improved.
Further: when the sweeper passes through a bumpy road section with two-level or three-level bump at a limited speed, the method further comprises the following steps:
firstly, an IMU inertial navigation component acquires an acceleration change data signal of a Z axis, and a vibration data processing unit adopts a stable Markov model (HMM) to judge and evaluate the data signal acquired by the IMU inertial navigation component to obtain first data of vibration of the sweeper;
secondly, acquiring vibration data of the sweeper by a vibration sensor, and obtaining second data of the sweeper vibration by a vibration data processing unit through weighted fusion calculation of a specific proportion;
calculating and estimating the final vibration data by the vibration data processing unit through a Kalman filtering algorithm on the first data and the second data to obtain vibration grades, wherein the vibration grades are divided into primary vibration, secondary vibration and tertiary vibration;
fourthly, the method comprises the following steps: and (3) judging whether the vibration grade obtained by the vibration data processing unit corresponds to the bumping grade identified in the step (2) or not by the main control module, if so, judging that the identification of the bumping road condition identification module is accurate, improving the confidence coefficient of the bumping road condition identification module, and if not, changing the bumping grade identified in the step (2) into the bumping grade corresponding to the vibration grade according to the vibration grade obtained by the vibration data processing unit, and feeding back the bumping grade to a cloud background system for training and correcting the road condition bumping identification module.
Further: the fourth step also comprises the following steps:
when the sweeper passes through a bumpy road section with a bumpy grade as a second bump at a limited speed, if the vibration grade obtained by the vibration data processing unit is a second-level vibration, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be accurate and fed back to the cloud background system, the confidence coefficient of the bumpy road condition recognition module is increased, if the vibration grade obtained by the vibration data processing unit is a first-level vibration, the bumpy grade of the bumpy road section is changed into a first-level bump, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be wrong, the recognition is fed back to the cloud background system, and the bumpy road condition recognition module is used;
when the sweeper passes through a bumpy road section with three-level bumpy jolt grade at a limited speed, if the vibration grade obtained by the vibration data processing unit is three-level vibration, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be accurate and fed back to the cloud background system, the confidence coefficient of the bumpy road condition recognition module is increased, if the vibration grade obtained by the vibration data processing unit is two-level vibration or the back camera detects that the cleaning effect is lower than a preset cleaning threshold value, the bumpy grade of the bumpy road section is changed into two-level bump, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be wrong, and the recognition is fed back to the cloud background system for the training and correction of the bumpy road.
Further: the step S6 further includes:
for the road section with class A bump, if GPS marking confirmation is carried out twice at the same position, the bump road section is marked as a fixed bump road section, if the sweeper passes the position twice continuously and does not recognize bump or the recognized bump is classified as non-class A, the bump road section is not marked as the fixed bump road section,
if the grade of the continuous three-time marking is first-level bump and the avoidance result is failure, the road section is avoided during global path planning, the previous marking is cleared and re-identified unless the road section is manually marked as a passable road section, and if the marking is first-level bump and the avoidance result is successful, or second-level bump and third-level bump, a local obstacle avoidance or deceleration strategy can be made in advance at the position during global path planning;
for a road section with class B bumps, if the grade of the continuous marking for three times is first-level bump and the avoidance result is failure, the road section is avoided during global path planning, unless the road section is manually marked to be a passable road section, the previous mark is cleared and re-identified, and if the mark is first-level bump and the avoidance structure is successful, or the mark is second-level bump and third-level bump, a local obstacle avoidance or deceleration strategy can be made in advance at the position during global path planning;
for the C-type bumpy road segment, if no bump information is identified at the position twice, the bump mark of the position is cancelled, otherwise, a local obstacle avoidance or deceleration strategy can be made at the position in advance during global path planning.
A self-adaptive safe driving system for bumpy road surfaces of an unmanned sweeper comprises a cloud background system and the sweeper, wherein a main control module is arranged in the sweeper and is connected with the main control module through communication equipment, the main control module is connected with a monocular camera, a binocular camera, a backward camera, a control module and a GPS module, a depth recognition unit and a vibration data processing unit are arranged in the main control module, the depth recognition unit is connected with an ultrasonic sensor, the vibration data processing unit is connected with an IMU (inertial navigation Unit) and a vibration sensor, the monocular camera, the binocular camera and the ultrasonic sensor are respectively arranged on the front side of the sweeper body, the IMU inertial navigation unit and the vibration sensor are arranged on the body chassis of the sweeper, the backward camera is arranged on the rear side of the sweeper body, the monocular camera, the binocular camera and the ultrasonic sensor form a bumpy road condition identification module.
Further: the monocular camera comprises a monocular identification unit, the monocular identification unit is connected with a camera, the binocular camera comprises a binocular identification unit, and the binocular identification unit is connected with two cameras.
The invention has the advantages of
Compared with the prior art, the shape, the size, the depth or the height of a bumpy road surface such as a speed bump, a hollow road surface, a concave-convex stripe cement brick and the like are identified by fusing the monocular camera, the binocular camera, the ultrasonic sensor, the IMU inertial navigation component and the plurality of vibration sensors, the bumpy road surface is classified into bumpy grades by combining the walking limit condition and the vibration condition of the sweeper and the sweeping effect of the bumpy road surface, the sweeper is controlled to make a driving strategy of avoiding, passing through at a reduced speed or passing directly aiming at the bumpy road surface with different bumpy grades, the driving strategy of the sweeper is adaptively adjusted according to the bumpy grade, the sweeper is ensured to safely pass through a bumpy road section, and the sweeping coverage and the operation effect of the sweeper are ensured to the maximum limit.
Drawings
FIG. 1 is a schematic diagram of an adaptive safe driving system for a bumpy road surface of an unmanned sweeper of the present invention;
FIG. 2 is a front view of the sweeper of the present invention;
FIG. 3 is a bottom view of the sweeper truck of the present invention;
fig. 4 is a side view of the sweeper truck of the present invention.
Description of reference numerals: the method comprises the following steps of 1-a cloud background system, 2-communication equipment, 3-a sweeper, 31-a sweeper body, 32-a chassis, 4-a monocular camera, 5-a binocular camera, 6-a back camera, 7-an ultrasonic sensor, 8-an IMU inertial navigation component and 9-a vibration sensor.
Detailed Description
Fig. 2 to 4 are schematic structural views of an embodiment of an unmanned sweeper bumpy road self-adaptive safe driving system provided by the invention, which includes a cloud background system 1 and a sweeper 3, the sweeper 3 is internally provided with a main control module, the cloud background system 1 is connected with the main control module through a communication device 2, the main control module is connected with a monocular camera 4, a binocular camera 5, a back camera 6, a control module and a GPS module, the main control module is internally provided with a depth recognition unit and a vibration data processing unit, the depth recognition unit is connected with an ultrasonic sensor 7, the vibration data processing unit is connected with an IMU inertial navigation component 8 and a vibration sensor 9, the monocular camera 4, the binocular camera 5 and the ultrasonic sensor 7 are respectively arranged on the front side of a sweeper body 31 of the sweeper 3, the IMU inertial navigation component 8 and the vibration sensor 9 are arranged on a sweeper body chassis 32 of the sweeper 3, the vibration sensor 9 is provided with threely, and 6 settings are in the automobile body 31 rear side of motor sweeper 3 dorsad to the camera, and the road conditions identification module of jolting is constituteed to monocular camera 4, two mesh cameras 5 and ultrasonic sensor 7, and monocular camera 4 includes monocular recognition unit, and monocular recognition unit is connected with a camera, and two mesh cameras 5 are including two mesh recognition unit, and two mesh recognition unit are connected with two cameras.
The back camera 6 is used for detecting the cleaning effect, the control module is used for controlling the sweeper to stop, decelerate, steer, avoid obstacles and the like according to the bumpy road condition, and the cloud background system 1 is used for marking bumpy road condition information, training and correcting the recognition model library, planning angles and paths of the sweeper operation task and the like.
Referring to fig. 1, the self-adaptive safe driving method for bumpy road surface of the unmanned sweeper comprises the following steps:
s1, acquiring an image of a road condition at a position 10-15 meters away from the advancing direction of the sweeper by a monocular camera, identifying and processing the image of the road condition by a monocular identification unit to identify the jolting type and the jolting section size of the jolting road section, obtaining the influence degree of the jolting road section on the operation of the sweeper according to the jolting type and the jolting size, and sending the jolting type and the jolting section size to a main control module;
s2, the main control module judges whether the influence degree of the bumpy road section on the operation of the sweeper reaches a preset safety passing threshold, if so, a deceleration instruction is sent to the control module to control the sweeper to decelerate to move forwards, if not, an original speed driving instruction is sent to the control module to control the sweeper to continue to move forwards at the original speed, and the information of the bump type and the size of the bumpy road section is reported to the cloud background system;
s3, when the sweeper moves forwards to a distance of 4-5 m from the bumpy road section, the images of the bumpy road section are collected by the binocular camera, the images of the bumpy road section are identified and processed by the binocular identification unit to obtain the bump type, the size and the depth of the bumpy road section, and the bump type, the size and the depth of the bumpy road section are sent to the main control module;
in this application, in step S1 monocular identification unit adopt monocular camera type of jolting training model right the road conditions image is discerned and is reachd the type of jolting, in step S3 binocular identification unit adopt binocular camera type of jolting training model right the image on the highway section of jolting is discerned and is reachd the type of jolting, wherein, the training method of monocular camera type of jolting training model and binocular camera type of jolting training model is the same, all includes following step:
acquiring a sample picture:
the sample comprises a pavement picture containing a sewer well cover, a deceleration strip, broken stones, branches and the like;
<2> feature marking:
marking bumpy elements of the sample picture in a manual mode, for example, marking sewer well covers, speed bumps, broken stones and the like in the picture, and manually classifying the bumpy elements, wherein the basic principle is whether the bumpy elements belong to permanence or temporary, and the elements to be included in each classification can be manually determined according to actual conditions;
<3> training model:
performing model training by taking the marked picture as an input element, and outputting a prediction model of a bump type according to a model training result, wherein the step of performing model training by taking the marked picture as the input element comprises the step of performing model training by a training method of KNN, decision tree, naive leaf Bayes, logistic regression and support vector machine;
<4> model identification:
test samples (without manual marks) are input into the prediction model, and the prediction model outputs a bumpy type recognition result according to previous training experience, so that classification and recognition of bumpy road segment classes are achieved.
The monocular identification unit processes the bumpy road section through the monocular camera calculation module to obtain the size of the bumpy road section, and the binocular identification unit processes the bumpy road section through the binocular camera calculation module to obtain the size of the bumpy road section.
S4, the main control module judges and verifies whether the bump type obtained by the recognition processing of the monocular camera and the binocular camera is consistent, if so, the bump type recognition of the monocular camera to the bump road section is accurate, the bump type recognition confidence coefficient of the monocular camera is improved, if not, the bump type obtained by the recognition of the binocular camera is taken as the standard, the recognition information is stored into a bump type training model of the monocular camera for the learning and correction of the monocular camera, the bump type recognition accuracy of the monocular camera is improved, in addition, the bump size of the road section obtained by the processing of the monocular camera is compared with the bump size obtained by the processing of the binocular camera, whether the difference between the two is in the size threshold range is judged, if so, the recognition of the monocular camera is judged to be accurate, the confidence coefficient of the monocular camera is improved, if not, the bump road section size obtained by the recognition of the binocular camera is taken as the standard, the size of the bumpy road section is fed back to the monocular camera calculating module for correction and learning of the monocular camera, and the recognition accuracy of the monocular camera is improved;
among them, the classification of the type of thrashing is as follows:
class a jounce: the road surface is characterized by comprising protruding speed bumps, sewer well covers, specific road surface material road sections (concave-convex stripe cement bricks and the like), specific terrain road sections (wave lines and the like) and other similar permanent bumpy road sections;
class B jounce: a repairable bumpy road section typified by a defective-type pothole on the road surface;
class C jounce: temporary bumpy road sections such as a broken stone road section and a residual branch road section;
s5, the main control module identifies the bump grade of the bump road section according to the bump road section depth, the bump type after judgment and verification and the bump road section size;
wherein, the first-order bumping: refers to a bumpy road condition seriously endangering the operation safety of the sweeper. Bumps which cannot be passed by the sweeper (pits, speed bumps and special road sections which exceed the design limit of the sweeper), bumps which are easy to cause the damage of key parts of the sweeper or the vibration of the whole sweeper (road conditions which can cause the failure of the sweeper due to the bump vibration);
secondary bumping: the road condition is bumpy, which may endanger the operation safety of the sweeper and seriously affect the sweeping effect of the sweeper, the sweeper must slowly pass through the road condition at a speed lower than a limited speed to ensure the bumping of the walking safety, and the sweeping effect of the sweeper is extremely poor;
third-order jounce: the cleaning device is a bumping condition which has no obvious influence on the operation safety of the sweeper even if the cleaning device lasts for a period of time (such as 20S), but has a certain influence on the cleaning effect;
in the application, according to the traffic capacity of hardware factors (tire type, tire diameter, motor power, motor torque, shaft distance) in the design of the sweeper, including obstacle crossing capacity, ditch crossing capacity, rollover prevention capacity and the like, a limit value and a preset danger threshold value when the sweeper passes through a bumpy road section are determined, wherein the limit value comprises the maximum limit height of the sweeper to a bump road section similar to a bump like a speed reducing belt and the like and the limit depth of the sweeper to concave bumpy roads such as pits and the like, and if the recognized bumpy road section is greater than the limit value, the sweeper is judged to be in first-level bump; if the identified bumpy road section is smaller than the limit value and larger than the preset danger threshold value, determining that the road section is bumpy for the second time; and if the bumpy road section is identified to be smaller than the preset danger threshold value, judging that the road section is bumpy in three levels.
S6, the main control module sends the driving strategy corresponding to the bumping grade to the control module according to the bumping grade identified in the step S5, the control module controls the sweeper to pass through the bumping road section according to the corresponding driving strategy, and the driving strategy is as follows:
when the identified bumping grade is first-level bumping, the sweeper starts an obstacle avoidance strategy, local path planning for bypassing a bumpy road section is carried out on the basis of original path planning, after obstacle avoidance is completed, the original planning is returned to continue executing tasks, information is reported to a cloud background system, and the reported information comprises bumping road condition information and avoidance results: successfully and actually shooting 5 images; if the road condition that the obstacle cannot be avoided is met, the sweeper stops, backs up, turns around and returns to the storehouse, information is reported to the cloud background system, and the reported information comprises jolting road condition information and avoidance results: 5 failed images are photographed;
when the identified bumping grade is two-stage bumping, the sweeper starts an obstacle avoidance strategy, local path planning for bypassing a bumpy road section is carried out on the basis of original path planning, after obstacle avoidance is completed, the original planning is returned again to continue executing tasks, information is reported to a cloud background system, and the reported information comprises bumpy road condition information and a passing mode: avoiding, bypassing and actually shooting 5 images; if the road condition that the obstacle cannot be avoided is met, the sweeper stops the cleaning function, decelerates to a limited speed, and reports information to the cloud background system, wherein the reported information comprises bumpy road condition information and a passing mode: 5 images are passed through and photographed at a reduced speed;
when the recognized bumping grade is three-level bumping, the sweeper decelerates to a limited speed, slowly passes through a bumping road section, the coverage rate of the cleaning operation is guaranteed to the maximum extent, the cleaning operation state is kept, information is reported to a cloud background system, and the reported information comprises bumping road condition information and a passing mode: 5 pictures are taken in real time after passing through the speed reduction;
the bumpy road condition information comprises the bumpy grade, the bumpy type and the GPS position of the bumpy road section.
When the sweeper slows down and passes through a bumpy road section with two-level bump and three-level bump and runs 40-60cm before the bumpy road section, the method further comprises the following steps:
(1): detecting the depth information of the bumpy road section through an ultrasonic sensor, and sending the depth information to a depth recognition unit for processing to obtain the depth of the bumpy road section;
(2): comparing the depth of the bumpy road section detected by the ultrasonic sensor with the depth of the bumpy road section processed by the binocular camera, judging whether the difference between the two is within a depth threshold value, if so, judging that the binocular camera is accurately identified, improving the confidence coefficient of the binocular camera, if not, regarding the information of the depth of the bumpy road section detected by the ultrasonic sensor as the standard, re-identifying the grade of the bumpy road section according to the bumpy road section depth detected by the ultrasonic sensor and the size of the bumpy road section after verification in step S4 to obtain the latest bumpy grade, judging whether the latest bumpy grade meets or is lower than the bumpy grade identified in step S5, if so, continuing to move the sweeper according to the driving strategy corresponding to the bumpy grade identified in step S5, and if not, emergently decelerating the sweeper according to the driving strategy corresponding to the latest bumpy grade, and the depth information of the bumpy road section detected by the ultrasonic sensor is fed back to the binocular camera calculation module for correction and learning of the binocular camera, so that the identification accuracy of the binocular camera is improved.
When the sweeper passes through a bumpy road section with two-level or three-level bump at a limited speed, the method further comprises the following steps:
firstly, an IMU inertial navigation component acquires an acceleration change data signal of a Z axis, and a vibration data processing unit adopts a stable Markov model (HMM) to judge and evaluate the data signal acquired by the IMU inertial navigation component to obtain first data of vibration of the sweeper;
secondly, acquiring vibration data of the sweeper by a vibration sensor, and obtaining second data of the sweeper vibration by a vibration data processing unit through weighted fusion calculation of a specific proportion;
calculating and estimating the first data and the second data by a vibration data processing unit through a Kalman filtering algorithm to obtain final vibration data to obtain vibration grades, wherein the vibration grades are divided into primary vibration, secondary vibration and tertiary vibration;
fourthly, the method comprises the following steps: and (3) judging whether the vibration grade obtained by the vibration data processing unit corresponds to the bumping grade identified in the step (2) or not by the main control module, if so, judging that the identification of the bumping road condition identification module is accurate, improving the confidence coefficient of the bumping road condition identification module, and if not, changing the bumping grade identified in the step (2) into the bumping grade corresponding to the vibration grade according to the vibration grade obtained by the vibration data processing unit, and feeding back the bumping grade to a cloud background system for training and correcting the road condition bumping identification module.
The fourth step also comprises:
when the sweeper passes through a bumpy road section with a bumpy grade as a second bump at a limited speed, if the vibration grade obtained by the vibration data processing unit is a second-level vibration, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be accurate and fed back to the cloud background system, the confidence coefficient of the bumpy road condition recognition module is increased, if the vibration grade obtained by the vibration data processing unit is a first-level vibration, the bumpy grade of the bumpy road section is changed into a first-level bump, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be wrong, the recognition is fed back to the cloud background system, and the bumpy road condition recognition module is used;
when the sweeper passes through a bumpy road section with three-level bumpy jolt grade at a limited speed, if the vibration grade obtained by the vibration data processing unit is three-level vibration, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be accurate and fed back to the cloud background system, the confidence coefficient of the bumpy road condition recognition module is increased, if the vibration grade obtained by the vibration data processing unit is two-level vibration or the back camera detects that the cleaning effect is lower than a preset cleaning threshold value, the bumpy grade of the bumpy road section is changed into two-level bump, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be wrong, and the recognition is fed back to the cloud background system for the training and correction of the bumpy road.
In the present application, the method for training the stable markov model is as follows:
the vibration data processing unit intercepts data signals acquired by the IMU inertial navigation component, decomposes the data signals by adopting an Empirical Mode Decomposition (EMD) method to generate the sum of a plurality of signal components, namely an eigenmode function (IMF), and codes the energy ratio corresponding to the eigenmode function component as a characteristic vector for judging the vibration degree caused by the level of a bumpy road section. The energy of the signal component is calculated as:
and N is the signal length, the energy of the signal component is normalized, a feature vector is constructed, and the calculation formula of the feature vector is as follows:
and E is the total signal energy, the feature vectors are coded to obtain a coding result, the coding result is input into a stable Markov model (HMM) to train the recognition of the vibration degrees caused by different bumping grades, and the vibration degree caused by each bumping grade trains an HMM model.
Step S6 further includes:
for the road section with class A bump, if GPS marking confirmation is carried out twice at the same position, the bump road section is marked as a fixed bump road section, if the sweeper passes the position twice continuously and does not recognize bump or the recognized bump is classified as non-class A, the bump road section is not marked as the fixed bump road section,
if the grade of the continuous three-time marking is first-level bump and the avoidance result is failure, the road section is avoided during global path planning, the previous marking is cleared and re-identified unless the road section is manually marked as a passable road section, and if the marking is first-level bump and the avoidance result is successful, or second-level bump and third-level bump, a local obstacle avoidance or deceleration strategy can be made in advance at the position during global path planning;
for a road section with class B bumps, if the grade of the continuous marking for three times is first-level bump and the avoidance result is failure, the road section is avoided during global path planning, unless the road section is manually marked to be a passable road section, the previous mark is cleared and re-identified, and if the mark is first-level bump and the avoidance structure is successful, or the mark is second-level bump and third-level bump, a local obstacle avoidance or deceleration strategy can be made in advance at the position during global path planning;
for the C-type bumpy road segment, if no bump information is identified at the position twice, the bump mark of the position is cancelled, otherwise, a local obstacle avoidance or deceleration strategy can be made at the position in advance during global path planning.
In conclusion, the shape, size, depth or height of bumpy road surfaces such as speed bumps, hollow and concave stripe cement bricks are identified by utilizing the monocular camera, the binocular camera, the ultrasonic sensor, the IMU inertial navigation component and the vibration sensors in a fusion mode, the bumpy road surfaces are classified according to the walking limit condition, the vibration condition and the bumpy road surface cleaning effect of the sweeper, the sweeper is controlled to make driving strategies of avoiding, passing through in a decelerating mode or passing through directly according to the bumpy road surfaces with different bumpy grades, the driving strategies of the sweeper are adjusted in a self-adaptive mode according to the bumpy grade, and the sweeping coverage and the operation effect of the sweeper are guaranteed to the maximum limit while the sweeper is guaranteed to pass through the bumpy road sections safely.
The above detailed description is specific to possible embodiments of the present invention, and the embodiments are not intended to limit the scope of the present invention, and all equivalent implementations or modifications that do not depart from the scope of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A self-adaptive safe driving method for bumpy road surfaces of an unmanned sweeper is characterized by comprising the following steps: the method comprises the following steps:
s1, acquiring an image of a road condition at a position 10-15 meters away from the advancing direction of the sweeper by a monocular camera, identifying and processing the image of the road condition by a monocular identification unit to identify the jolting type and the jolting section size of the jolting road section, obtaining the influence degree of the jolting road section on the operation of the sweeper according to the jolting type and the jolting section size, and sending the influence degree of the jolting type, the jolting section size and the jolting road section to a main control module;
s2, the main control module judges whether the influence degree of the bumpy road section on the operation of the sweeper reaches a preset safety passing threshold, if so, a deceleration instruction is sent to the control module to control the sweeper to decelerate to move forwards, if not, an original speed driving instruction is sent to the control module to control the sweeper to continue to move forwards at the original speed, and the information of the bump type and the size of the bumpy road section is reported to the cloud background system;
s3, when the sweeper moves forwards to a distance of 4-5 m from the bumpy road section, the images of the bumpy road section are collected by the binocular camera, the images of the bumpy road section are identified and processed by the binocular identification unit to obtain the bump type, the size and the depth of the bumpy road section, and the bump type, the size and the depth of the bumpy road section are sent to the main control module;
s4, the main control module judges and verifies whether the bump type obtained by the recognition processing of the monocular camera and the binocular camera is consistent, if so, the bump type recognition of the monocular camera to a bump road section is accurate, the bump type recognition confidence coefficient of the monocular camera is improved, if not, the bump type obtained by the recognition of the binocular camera is taken as the standard, the recognition information is stored in the monocular camera for the learning and correction of the monocular camera, the bump type recognition accuracy of the monocular camera is improved, in addition, the bump road section size obtained by the processing of the monocular camera is compared with the bump road section size obtained by the processing of the binocular camera, whether the difference between the two is in the size threshold range is judged, if so, the recognition of the monocular camera is judged to be accurate, the confidence coefficient of the monocular camera is improved, if not, the bump road section size obtained by the recognition of the binocular camera is taken as the standard, and the bump road section size is fed back to the monocular camera, the monocular camera is used for correction and learning, and the recognition accuracy of the monocular camera is improved;
s5, the main control module identifies the bump grade of the bump road section according to the bump road section depth measured by the binocular camera and the bump type and the bump road section size judged and verified in the step S4;
and S6, the main control module sends the driving strategy corresponding to the bumping grade to the control module according to the bumping grade identified in the step S5, and the control module controls the sweeper to pass through the bumping road section according to the corresponding driving strategy.
2. The self-adaptive safe driving method for the bumpy road surface of the unmanned sweeper according to claim 1, wherein the self-adaptive safe driving method comprises the following steps: the classification of the bump types is as follows:
class a jounce: the system comprises a convex speed bump, a sewer well cover, a road section made of a specific road surface material and a road section with specific terrain, wherein the road section is permanently bumpy;
class B jounce: a repairable bumpy road section typified by a defective-type pothole on the road surface;
class C jounce: the temporary bumpy road sections of the broken stone road section and the residual branch road section.
3. The self-adaptive safe driving method for the bumpy road surface of the unmanned sweeper according to claim 2, wherein the self-adaptive safe driving method comprises the following steps: the bump rating is classified as follows:
the first-level bump is a bumpy road condition seriously endangering the running safety of the sweeper; bumping which the sweeper cannot pass through and bumping which is easy to cause the damage of key parts of the sweeper or the vibration of the whole sweeper;
secondary bumping: the road condition is bumpy, which may endanger the operation safety of the sweeper and seriously affect the sweeping effect of the sweeper, the sweeper must slowly pass through the road condition at a speed lower than a limited speed to ensure the bumping of the walking safety, and the sweeping effect of the sweeper is extremely poor;
third-order jounce: the cleaning vehicle is a bumping condition which has no obvious influence on the operation safety of the cleaning vehicle even if the cleaning vehicle lasts for a period of time, but has a certain influence on the cleaning effect.
4. The self-adaptive safe driving method for the bumpy road surface of the unmanned sweeper according to claim 3, wherein the self-adaptive safe driving method comprises the following steps: the driving strategy described in step S6 is as follows:
when the identified bumping grade is first-level bumping, the sweeper starts an obstacle avoidance strategy, local path planning for bypassing a bumpy road section is carried out on the basis of original path planning, after obstacle avoidance is completed, the original planning is returned to continue executing tasks, information is reported to a cloud background system, information of bumpy road conditions and avoidance results are reported: successfully and actually shooting 5 images; if the road condition that the obstacle cannot be avoided is met, the sweeper stops, backs up, turns around and returns to the storehouse, information is reported to the cloud background system, and the reported information comprises jolting road condition information and avoidance results: 5 failed images are photographed;
when the identified bumping grade is two-stage bumping, the sweeper starts an obstacle avoidance strategy, local path planning for bypassing a bumpy road section is carried out on the basis of original path planning, after obstacle avoidance is completed, the original planning is returned again to continue executing tasks, information is reported to a cloud background system, and the reported information comprises bumpy road condition information and a passing mode: avoiding, bypassing and actually shooting 5 images; if the road condition that the obstacle cannot be avoided is met, the sweeper stops the cleaning function, decelerates to a limited speed, and reports information to the cloud background system, wherein the reported information comprises bumpy road condition information and a passing mode: 5 images are passed through and photographed at a reduced speed;
when the recognized bumping grade is three-level bumping, the sweeper decelerates to a limited speed, slowly passes through a bumping road section, the coverage rate of the cleaning operation is guaranteed to the maximum extent, the cleaning operation state is kept, information is reported to a cloud background system, and the reported information comprises bumping road condition information and a passing mode: 5 pictures are taken in real time after passing through the speed reduction;
the bumpy road condition information comprises the bumpy grade, the bumpy type and the GPS position of the bumpy road section.
5. The self-adaptive safe driving method for the bumpy road surface of the unmanned sweeper according to claim 4, wherein the self-adaptive safe driving method comprises the following steps: when the sweeper slows down and passes through a bumpy road section with two-level bump and three-level bump and runs 40-60cm before the bumpy road section, the method further comprises the following steps:
(1): detecting the depth information of the bumpy road section through an ultrasonic sensor, and sending the depth information to a depth recognition unit for processing to obtain the depth of the bumpy road section;
(2): comparing the depth of the bumpy road section detected by the ultrasonic sensor with the depth of the bumpy road section processed by the binocular camera, judging whether the difference between the two is within a depth threshold value, if so, judging that the binocular camera is accurately identified, improving the confidence coefficient of the binocular camera, if not, regarding the information of the depth of the bumpy road section detected by the ultrasonic sensor as the standard, re-identifying the grade of the bumpy road section according to the bumpy road section depth detected by the ultrasonic sensor and the size of the bumpy road section after verification in step S4 to obtain the latest bumpy grade, judging whether the latest bumpy grade meets or is lower than the bumpy grade identified in step S5, if so, continuing to move the sweeper according to the driving strategy corresponding to the bumpy grade identified in step S5, and if not, emergently decelerating the sweeper according to the driving strategy corresponding to the latest bumpy grade, and the depth information of the bumpy road section detected by the ultrasonic sensor is fed back to the binocular camera for correction and learning of the binocular camera, so that the identification accuracy of the binocular camera is improved.
6. The self-adaptive safe driving method for the bumpy road surface of the unmanned sweeper according to claim 5, wherein the self-adaptive safe driving method comprises the following steps: when the sweeper passes through a bumpy road section with two-level or three-level bump at a limited speed, the method further comprises the following steps:
firstly, an IMU inertial navigation component acquires an acceleration change data signal of a Z axis, and a vibration data processing unit adopts a stable Markov model (HMM) to judge and evaluate the data signal acquired by the IMU inertial navigation component to obtain first data of vibration of the sweeper;
secondly, acquiring vibration data of the sweeper by using a vibration sensor, and acquiring second data of the sweeper vibration by using a vibration data processing unit through weighted fusion calculation;
calculating and estimating the final vibration data by the vibration data processing unit through a Kalman filtering algorithm on the first data and the second data to obtain vibration grades, wherein the vibration grades are divided into primary vibration, secondary vibration and tertiary vibration;
fourthly, the method comprises the following steps: and (3) judging whether the vibration grade obtained by the vibration data processing unit corresponds to the bumping grade identified in the step (2) or not by the main control module, if so, judging that the identification of the bumping road condition identification module is accurate, improving the confidence coefficient of the bumping road condition identification module, and if not, changing the bumping grade identified in the step (2) into the bumping grade corresponding to the vibration grade according to the vibration grade obtained by the vibration data processing unit, and feeding back the bumping grade to the cloud background system for training and correcting the road condition bumping identification module.
7. The self-adaptive safe driving method for the bumpy road surface of the unmanned sweeper as claimed in claim 6, wherein the self-adaptive safe driving method comprises the following steps: the fourth step also comprises the following steps:
when the sweeper passes through a bumpy road section with a bumpy grade as a second bump at a limited speed, if the vibration grade obtained by the vibration data processing unit is a second-level vibration, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be accurate and fed back to the cloud background system, the confidence coefficient of the bumpy road condition recognition module is increased, if the vibration grade obtained by the vibration data processing unit is a first-level vibration, the bumpy grade of the bumpy road section is changed into a first-level bump, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be wrong, the recognition is fed back to the cloud background system, and the bumpy road condition recognition module is used;
when the sweeper passes through a bumpy road section with three-level bumpy jolt grade at a limited speed, if the vibration grade obtained by the vibration data processing unit is three-level vibration, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be accurate and fed back to the cloud background system, the confidence coefficient of the bumpy road condition recognition module is increased, if the vibration grade obtained by the vibration data processing unit is two-level vibration or the back camera detects that the cleaning effect is lower than a preset cleaning threshold value, the bumpy grade of the bumpy road section is changed into two-level bump, the recognition of the bumpy road condition recognition module on the bumpy grade is judged to be wrong, and the recognition is fed back to the cloud background system for the training and correction of the bumpy road.
8. The self-adaptive safe driving method for the bumpy road surface of the unmanned sweeper according to claim 7, wherein the self-adaptive safe driving method comprises the following steps: the step S6 further includes:
for the road section with class A bump, if GPS marking confirmation is carried out twice at the same position, the bump road section is marked as a fixed bump road section, if the sweeper passes the position twice continuously and does not recognize bump or the recognized bump is classified as non-class A, the bump road section is not marked as the fixed bump road section,
if the grade of the continuous three-time marking is first-level bump and the avoidance result is failure, the road section is avoided during global path planning, the previous marking is cleared and re-identified unless the road section is manually marked as a passable road section, and if the marking is first-level bump and the avoidance result is successful, or second-level bump and third-level bump, a local obstacle avoidance or deceleration strategy can be made in advance at the position during global path planning;
for a road section with class B bumps, if the grade of the continuous marking for three times is first-level bump and the avoidance result is failure, the road section is avoided during global path planning, unless the road section is manually marked to be a passable road section, the previous mark is cleared and re-identified, and if the mark is first-level bump and the avoidance structure is successful, or the mark is second-level bump and third-level bump, a local obstacle avoidance or deceleration strategy can be made in advance at the position during global path planning;
for the C-type bumpy road segment, if no bump information is identified at the position twice, the bump mark of the position is cancelled, otherwise, a local obstacle avoidance or deceleration strategy can be made at the position in advance during global path planning.
9. An unmanned sweeping vehicle bumpy road surface adaptive safe driving system applying the safe driving method according to any one of claims 1 to 8, characterized in that: comprises a cloud background system and a sweeper, wherein the sweeper is internally provided with a main control module, the cloud background system is connected with the main control module through communication equipment, the main control module is connected with the monocular camera, the binocular camera, the back camera, the control module and the GPS module, the master control module is internally provided with a depth recognition unit and a vibration data processing unit, the depth recognition unit is connected with an ultrasonic sensor, the vibration data processing unit is connected with the IMU inertial navigation component and the vibration sensor, the monocular camera, the binocular camera and the ultrasonic sensor are respectively arranged on the front side of the sweeper body, the IMU inertial navigation component and the vibration sensor are arranged on a vehicle body chassis of the sweeper, the backward camera is arranged on the rear side of the sweeper body, and the monocular camera, the binocular camera and the ultrasonic sensor form a bumpy road condition identification module.
10. The self-adaptive safe driving system for the bumpy road surface of the unmanned sweeping vehicle as claimed in claim 9, wherein: the monocular camera comprises a monocular identification unit, the monocular identification unit is connected with a camera, the binocular camera comprises a binocular identification unit, and the binocular identification unit is connected with two cameras.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010855733.9A CN111708372B (en) | 2020-08-24 | 2020-08-24 | Self-adaptive safe driving method and system for bumpy road surface of unmanned sweeper |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010855733.9A CN111708372B (en) | 2020-08-24 | 2020-08-24 | Self-adaptive safe driving method and system for bumpy road surface of unmanned sweeper |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111708372A CN111708372A (en) | 2020-09-25 |
| CN111708372B true CN111708372B (en) | 2020-12-29 |
Family
ID=72547352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010855733.9A Active CN111708372B (en) | 2020-08-24 | 2020-08-24 | Self-adaptive safe driving method and system for bumpy road surface of unmanned sweeper |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111708372B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2783780C1 (en) * | 2021-10-04 | 2022-11-17 | Общество с ограниченной ответственностью «ЭвоКарго» | Method for adaptive control of movement characteristics of highly automated vehicle depending on level of uncertainty in assessment of surrounding situation |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113311820B (en) * | 2021-04-03 | 2022-11-08 | 联友智连科技有限公司 | Unmanned vehicle automatic maneuvering system and method based on big data |
| CN113392729A (en) * | 2021-05-28 | 2021-09-14 | 北京百度网讯科技有限公司 | Method and device for generating bump data |
| CN113384772A (en) * | 2021-08-02 | 2021-09-14 | 中南大学湘雅医院 | Pancreas washing unit |
| CN113771573B (en) * | 2021-09-22 | 2023-06-13 | 北京车和家信息技术有限公司 | Vehicle suspension control method and device based on identification road surface information |
| WO2023059221A1 (en) * | 2021-10-04 | 2023-04-13 | Общество с ограниченной ответственностью "ЭвоКарго" | Method of controlling vehicle driving characteristics |
| CN113762428A (en) * | 2021-11-10 | 2021-12-07 | 北京中科慧眼科技有限公司 | Road surface bumping degree grade classification method and system |
| CN115230718A (en) * | 2021-11-23 | 2022-10-25 | 上海仙途智能科技有限公司 | Control method and system for unmanned sweeper and readable storage medium |
| CN114347903A (en) * | 2021-12-30 | 2022-04-15 | 中交二公局第六工程有限公司 | Prefabricated comprehensive pipe rack transportation bump early warning system |
| CN114475573B (en) * | 2022-02-14 | 2022-11-15 | 科大国创合肥智能汽车科技有限公司 | Fluctuating road condition identification and vehicle control method based on V2X and vision fusion |
| CN114735022B (en) * | 2022-03-15 | 2024-08-23 | 江西赛特智能科技有限公司 | Method and system for preventing unmanned sweeper from blocking in narrow channel and sweeper |
| CN114872029B (en) * | 2022-06-09 | 2024-02-02 | 深圳市巨龙创视科技有限公司 | Robot vision recognition system |
| CN115320307B (en) * | 2022-10-17 | 2023-01-20 | 浙江孔辉汽车科技有限公司 | Vibration damping method and device for vehicle, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107478238A (en) * | 2017-07-31 | 2017-12-15 | 中北智杰科技（北京）有限公司 | A kind of new-energy automobile Unmanned Systems and control method |
| CN109797691A (en) * | 2019-01-29 | 2019-05-24 | 浙江联运知慧科技有限公司 | Unmanned sweeper and its travelling-crane method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7744122B2 (en) * | 1995-12-12 | 2010-06-29 | Automotive Technologies International, Inc. | Driver side aspirated airbags |
| JP2001134320A (en) * | 1999-11-01 | 2001-05-18 | Honda Motor Co Ltd | Lane follow-up controller |
| KR101454824B1 (en) * | 2013-04-03 | 2014-11-03 | 국방과학연구소 | System and Method for estimating positions of an autonomous mobile vehicle |
| CN108221603B (en) * | 2018-01-08 | 2019-08-09 | 重庆大学 | A kind of road surface three-dimensional information detection device, the method and system of road |
| CN108691309A (en) * | 2018-02-24 | 2018-10-23 | 国电大渡河流域水电开发有限公司 | One kind rolling machine automatic drive system |
-
2020
- 2020-08-24 CN CN202010855733.9A patent/CN111708372B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107478238A (en) * | 2017-07-31 | 2017-12-15 | 中北智杰科技（北京）有限公司 | A kind of new-energy automobile Unmanned Systems and control method |
| CN109797691A (en) * | 2019-01-29 | 2019-05-24 | 浙江联运知慧科技有限公司 | Unmanned sweeper and its travelling-crane method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2783780C1 (en) * | 2021-10-04 | 2022-11-17 | Общество с ограниченной ответственностью «ЭвоКарго» | Method for adaptive control of movement characteristics of highly automated vehicle depending on level of uncertainty in assessment of surrounding situation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111708372A (en) | 2020-09-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111708372B (en) | Self-adaptive safe driving method and system for bumpy road surface of unmanned sweeper | |
| US10713504B2 (en) | Estimating friction based on image data | |
| US11008008B1 (en) | Modifying a vehicle state based on the presence of a special-purpose vehicle | |
| US11938967B2 (en) | Preparing autonomous vehicles for turns | |
| EP3548845B1 (en) | Navigation based on vehicle activity | |
| JP7466396B2 (en) | Vehicle control device | |
| CN103718225B (en) | Driving assist system and traveling householder method | |
| CN109131067B (en) | Tripod self-propelled vehicle and obstacle avoidance method thereof | |
| KR20200016949A (en) | FUSION FRAMEWORK and BATCH ALIGNMENT of Navigation Information for Autonomous Driving | |
| US10928832B2 (en) | Impedance-based motion control for autonomous vehicles | |
| WO2016080100A1 (en) | Drive control system | |
| EP3544856A1 (en) | Determining a road surface characteristic | |
| KR20200127219A (en) | Navigating a vehicle based on a detected barrier | |
| US9360332B2 (en) | Method for determining a course of a traffic lane for a vehicle | |
| US11535271B2 (en) | Methods and systems for monitoring vehicle motion with driver safety alerts | |
| US10688841B1 (en) | Expanding sensor domain coverage using differential active suspension | |
| CA3155591C (en) | Functional safety in autonomous driving | |
| US20220063604A1 (en) | Vehicle travel control device | |
| US20210339777A1 (en) | System and method for controlling one or more vehicles with one or more controlled vehicles | |
| US11854229B2 (en) | Object localization for autonomous driving by visual tracking and image reprojection | |
| US20220366175A1 (en) | Long-range object detection, localization, tracking and classification for autonomous vehicles | |
| Lee et al. | Ods-bot: Mobile robot navigation for outdoor delivery services | |
| Chipka et al. | Estimation and navigation methods with limited information for autonomous urban driving | |
| WO2022050006A1 (en) | Image processing device | |
| CN117999204A (en) | Vehicle control device, vehicle control method, and vehicle control system |
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 | ||
| CP02 | Change in the address of a patent holder |
Address after: 510000 201, building a, No.19 nanxiangsan Road, Huangpu District, Guangzhou City, Guangdong Province Patentee after: GUANGZHOU SAITE INTELLIGENT TECHNOLOGY Co.,Ltd. Address before: 510000 Room 303, 36 Kaitai Avenue, Huangpu District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU SAITE INTELLIGENT TECHNOLOGY Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder |