CN119148722B - A mobile charging robot obstacle avoidance method and system based on multi-sensor fusion - Google Patents
A mobile charging robot obstacle avoidance method and system based on multi-sensor fusion Download PDFInfo
- Publication number
- CN119148722B CN119148722B CN202411625856.8A CN202411625856A CN119148722B CN 119148722 B CN119148722 B CN 119148722B CN 202411625856 A CN202411625856 A CN 202411625856A CN 119148722 B CN119148722 B CN 119148722B
- Authority
- CN
- China
- Prior art keywords
- obstacle
- obstacle avoidance
- grade
- environmental data
- environment
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000004927 fusion Effects 0.000 title claims abstract description 24
- 230000007613 environmental effect Effects 0.000 claims abstract description 159
- 230000006870 function Effects 0.000 claims description 69
- 230000008447 perception Effects 0.000 claims description 46
- 230000008859 change Effects 0.000 claims description 35
- 230000003068 static effect Effects 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 9
- 230000004888 barrier function Effects 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
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
- G05D1/40—Control within particular dimensions
- G05D1/43—Control of position or course in two dimensions
-
- 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
- G05D1/20—Control system inputs
- G05D1/24—Arrangements for determining position or orientation
- G05D1/242—Means based on the reflection of waves generated by 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
- G05D1/20—Control system inputs
- G05D1/24—Arrangements for determining position or orientation
- G05D1/243—Means capturing signals occurring naturally from the environment, e.g. ambient optical, acoustic, gravitational or magnetic signals
-
- 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
- G05D1/20—Control system inputs
- G05D1/24—Arrangements for determining position or orientation
- G05D1/246—Arrangements for determining position or orientation using environment maps, e.g. simultaneous localisation and mapping [SLAM]
-
- 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
- G05D1/60—Intended control result
- G05D1/617—Safety or protection, e.g. defining protection zones around obstacles or avoiding hazards
- G05D1/622—Obstacle avoidance
- G05D1/633—Dynamic obstacles
-
- 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
- G05D1/60—Intended control result
- G05D1/644—Optimisation of travel parameters, e.g. of energy consumption, journey time or distance
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a mobile charging robot obstacle avoidance method and system based on multi-sensor fusion, which belong to the technical field of robots and comprise the steps of acquiring surrounding environment information in real time to obtain first environment data, judging an obstacle avoidance grade according to the first environment data, detecting surrounding obstacles through emission radio waves to perform rough obstacle avoidance if the obstacle avoidance grade is primary, acquiring second environment data through a multi-sensor and predicting an obstacle movement track according to the second environment data to plan an obstacle avoidance path to perform obstacle avoidance if the obstacle avoidance grade is intermediate, and acquiring third environment data through unmanned aerial vehicle real-time scanning and combining the second environment data to construct a real-time dynamic environment prediction model of the next moment if the obstacle avoidance grade is high-grade, and dynamically planning the obstacle avoidance path after the next moment to perform obstacle avoidance. The invention adopts the strategy of collecting the environmental information in real time and judging the obstacle avoidance grade, can take obstacle avoidance measures with different precision according to different grades, and improves the accuracy and efficiency of obstacle avoidance.
Description
Technical Field
The invention relates to the technical field of robots, in particular to a mobile charging robot obstacle avoidance method and system based on multi-sensor fusion.
Background
With the rapid development of technology, mobile charging robots are increasingly used in various places such as shopping centers, airports, hospitals and the like. The robots not only provide convenient charging service for customers, but also promote the overall intelligent level of places. However, in practical applications, mobile charging robots face complex obstacle avoidance challenges. Particularly, in an environment with large traffic and many obstacles, how to ensure that a robot can safely and efficiently shuttle between people and the obstacles becomes a problem to be solved urgently.
Traditional obstacle avoidance techniques lack the ability to flexibly adjust obstacle avoidance strategies according to different environmental complexities. In low complexity environments, simple sensors such as infrared sensors may provide sufficient environmental information. However, in a highly complex, multiple dynamic obstacle environment, such single sensors and corresponding obstacle avoidance algorithms do not meet the safety and efficiency requirements. In order to cope with the problems, some researches propose to use multi-sensor data to perform comprehensive judgment, but the problems of high computational complexity and insufficient real-time performance still exist, particularly in emergency situations, the fast obstacle avoidance decision making can be limited, and in addition, a mechanism for accurately predicting the future obstacle path is also lacked, which is particularly important in a high dynamic environment. Therefore, it is necessary to provide an obstacle avoidance method capable of adjusting an obstacle avoidance strategy in real time and having high-precision path planning capability, so as to ensure efficient operation of the mobile robot in a complex environment.
Disclosure of Invention
In order to solve the problems, the invention provides the obstacle avoidance method and the system for the mobile charging robot based on multi-sensor fusion, which adopt a strategy of collecting environmental information in real time and judging the obstacle avoidance grade, can take obstacle avoidance measures with different precision according to different grades, and improve the accuracy and the efficiency of obstacle avoidance.
The above object can be achieved by the following scheme:
A mobile charging robot obstacle avoidance method based on multi-sensor fusion comprises the steps of collecting surrounding environment information in real time to obtain first environment data, judging an obstacle avoidance grade according to the first environment data, detecting surrounding obstacles through emitting radio waves to conduct rough obstacle avoidance if the obstacle avoidance grade is primary, collecting second environment data through multiple sensors, predicting obstacle movement tracks according to the second environment data, planning an obstacle avoidance path to conduct obstacle avoidance, and conducting real-time scanning through an unmanned aerial vehicle to obtain third environment data if the obstacle avoidance grade is high-grade, constructing a real-time dynamic environment prediction model at the next moment according to the second environment data and the third environment data, and conducting obstacle avoidance after the next moment by combining the real-time dynamic environment prediction model at the next moment.
Optionally, judging the obstacle avoidance grade according to the first environmental data comprises judging whether an obstacle exists around according to the first environmental data, if no obstacle exists around, not starting an obstacle avoidance function, and if the obstacle exists around, judging the obstacle avoidance grade.
Optionally, if the surrounding obstacle exists, judging the obstacle avoidance grade further comprises judging the state of the obstacle according to the first environmental data, judging the obstacle avoidance grade as a primary if the obstacle is static, and judging the obstacle avoidance grade according to the direction change times, the number and the speed of the obstacle if the obstacle is dynamic.
Optionally, the step of judging the obstacle avoidance grade according to the number, the number and the speed of the direction change of the obstacle comprises the steps of setting a first weight factor for the number of the direction change of the obstacleSetting a second weight factor for the speed of the obstacleScoring the number, quantity and speed of direction change of the obstacle, and constructing a scoring matrix of the number and speed of direction change of the obstacleWherein each row represents an obstacle, each column represents an evaluation condition (including the number of direction changes and the speed), and the scoring matrixThe structure of (2) is as follows:
,
in the formula, Is the firstA direction change number score value for each obstacle; Is the first Calculating the direction change times weighted score and the speed weighted score of each obstacle, and adding the direction change times weighted score and the speed weighted score of each obstacle to obtain the weighted total score of each obstacleThere is
;
Adding the weighted total score value of all the barriers to the score value of the number of barriers to obtain an actual score valueFor the actual scoring valueHas the following components
,
In the formula,Is the number of obstacles; And judging the obstacle avoidance grade according to the actual grading value and a first threshold value.
Optionally, the scoring of the number, the number and the speed of the direction change of the obstacle comprises setting a second threshold of the number of the direction change of the obstacle, setting a third threshold of the speed of the obstacle, setting a fourth threshold of the number of the obstacle, obtaining a first score if the number of the direction change of the obstacle is smaller than the second threshold, otherwise obtaining a second score, obtaining a third score if the speed of the obstacle is smaller than the third threshold, otherwise obtaining a fourth score, and obtaining a fifth score if the number of the obstacle is smaller than the fourth threshold, otherwise obtaining a sixth score.
Optionally, the step of judging the obstacle avoidance grade according to the actual grading value and the first threshold value comprises the steps of judging the actual grading value and the first threshold value, judging the obstacle avoidance grade as a middle grade if the actual grading value is smaller than the first threshold value, and judging the obstacle avoidance grade as an advanced grade if the actual grading value is larger than or equal to the first threshold value.
Optionally, the predicting the movement track of the obstacle according to the second environmental data, planning the obstacle avoidance path to avoid the obstacle includes establishing an environmental perception function, obtaining a first environmental perception result according to the second environmental data and the environmental perception function, and obtaining the first environmental perception result according to the first environmental perception resultThere is
,
In the formula,As a function of the perception of the environment,Is thatEstablishing a motion prediction function, collecting historical motion data of the obstacle, combining the environment sensing result, obtaining a first obstacle motion trail prediction result through the motion prediction function, and obtaining the first obstacle motion trail prediction resultThere is
,
In the formula,Representation ofA first obstacle movement trajectory prediction result of the moment,The motion prediction function is represented by a function of motion,Representation ofEstablishing a first path planning function, obtaining a first path planning result according to the obstacle movement track prediction result and the first path planning function, and obtaining the first path planning result according to the first path planning resultThere is
,
In the formula,The result of the first path planning is indicated,And planning a robot motion path according to the first path planning result to realize obstacle avoidance.
Optionally, the constructing a real-time dynamic environment prediction model of the next moment according to the second environment data and the third environment data, combining the real-time dynamic environment prediction model of the next moment, and dynamically planning the obstacle avoidance path after the next moment to avoid the obstacle comprises calculating the related parameters of the environment perception of the next moment according to the environment perception function, the second environment data and the third environment data to obtain a second environment perception resultCalculating the obstacle motion trail after the next moment according to the second environment sensing result, the motion prediction function and the first obstacle motion trail prediction result to obtain a second obstacle motion trail prediction result, and regarding the second obstacle motion trail prediction resultThere is
,
In the formula,Representation ofEstablishing a real-time dynamic environment prediction model of the next moment according to the second environment sensing result and the second obstacle motion trail prediction result, and establishing the real-time dynamic environment prediction model of the next moment according to the second environment sensing result and the second obstacle motion trail prediction resultThere is
,
In the formula,Representing a function for constructing a real-time dynamic environment prediction model at the next moment, acquiring obstacle position coordinate information according to the third environment data, establishing a second path planning function, calculating a planned path after the next moment to obtain a second path planning result according to the second path planning function, the obstacle position coordinate information, the real-time dynamic environment model and the second obstacle movement track prediction result, and obtaining the second path planning result for the second path planning resultThere is
,
In the formula,A function is planned for the second path and,Is obstacle position coordinate information.
Optionally, the calculating the environmental perception related parameters of the next moment to obtain a second environmental perception result according to the environmental perception function, the second environmental data and the third environmental data comprises performing target matching and time-space alignment on the second environmental data and the third environmental data to obtain fourth environmental data, establishing a prediction model for describing the change of the fourth environmental data along with time, calculating a predicted value of the fourth environmental data at the current moment according to the fourth environmental data at the previous moment through the prediction model, training the prediction model according to the fourth environmental data at the current moment and the predicted value to obtain a stable prediction model, calculating the fourth environmental data at the next moment to obtain fifth environmental data according to the stable prediction model prediction and the fourth environmental data, calculating the environmental perception related parameters at the next moment to obtain the second environmental perception result according to the environmental perception function and the fifth environmental data, and obtaining the second environmental perception result for the second environmental perception resultThere are;
,
in the formula, Is the fifth environmental data.
Based on the same inventive concept, the invention further provides a mobile charging robot obstacle avoidance system based on multi-sensor fusion, which comprises an environment data acquisition module, an obstacle avoidance grade decision module, an obstacle avoidance strategy execution module and a motion control module, wherein the environment data acquisition module is used for acquiring surrounding environment information in real time to obtain first environment data, the obstacle avoidance grade decision module is used for judging an obstacle avoidance grade according to the first environment data, the obstacle avoidance strategy execution module is used for detecting surrounding obstacles through a radio wave emission module if the obstacle avoidance grade is primary, the obstacle avoidance strategy execution module is also used for acquiring second environment data through a plurality of sensors and predicting an obstacle motion track according to the second environment data when the obstacle avoidance grade is secondary, planning an obstacle avoidance path, the obstacle avoidance strategy execution module is also used for acquiring third environment data through unmanned aerial vehicle real-time scanning and constructing a real-time dynamic environment prediction model according to the second environment data and the third environment data, and combining the real-time dynamic environment prediction model, and the motion control module is used for controlling the movement of the robot according to control.
Compared with the prior art, the invention has the following advantages:
1. The invention can adopt obstacle avoidance strategies with different precision by collecting surrounding environment information in real time and judging obstacle avoidance grades according to the complexity of the information and the dynamic property of the obstacle, and can carry out rough detection by transmitting radio waves in a primary obstacle avoidance stage;
2. The method can adapt to environments with different complexity, can adopt a simpler obstacle avoidance strategy to save calculation resources in environments with fewer obstacles or slower moving speed, and can rapidly switch to a higher-level obstacle avoidance mode in scenes with dense obstacles or faster moving speed to ensure safe passing;
3. According to the multi-sensor fusion obstacle avoidance method, the motion state and the surrounding environment state of the obstacle at the current moment are predicted in advance, so that the motion state and the surrounding environment information of the obstacle at the next moment at the current moment are predicted, the robot can avoid the obstacle more intelligently, the collision and the dead time are reduced, and the efficient obstacle avoidance in a complex environment is realized;
4. the invention adopts multi-sensor fusion, can fully utilize the complementary advantages of different sensors, improves the sensing capability and the adaptability of the system to environmental changes, and can still perform reliable obstacle avoidance operation by depending on the information provided by other sensors even if some sensors have faults or data anomalies.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a mobile charging robot obstacle avoidance method based on multi-sensor fusion in an embodiment of the invention.
Fig. 2 is a flowchart of an implementation of a mobile charging robot obstacle avoidance method based on multi-sensor fusion according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of a mobile charging robot obstacle avoidance system based on multi-sensor fusion according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, an embodiment of the invention provides a mobile charging robot obstacle avoidance method based on multi-sensor fusion, and by adopting a strategy of collecting environmental information in real time and judging an obstacle avoidance grade, obstacle avoidance measures with different precision can be adopted according to different grades, so that the accuracy and efficiency of obstacle avoidance are improved.
The method of this embodiment specifically includes:
collecting surrounding environment information in real time to obtain first environment data;
judging an obstacle avoidance grade according to the first environmental data;
if the obstacle avoidance grade is primary, detecting surrounding obstacles by transmitting radio waves to perform rough obstacle avoidance;
If the obstacle avoidance grade is a middle grade, acquiring second environmental data through a plurality of sensors, predicting an obstacle movement track according to the second environmental data, and planning an obstacle avoidance path to avoid an obstacle;
And if the obstacle avoidance grade is advanced, acquiring third environmental data through real-time scanning of the unmanned aerial vehicle, constructing a real-time dynamic environmental prediction model of the next moment according to the second environmental data and the third environmental data, and dynamically planning an obstacle avoidance path after the next moment to avoid the obstacle by combining the real-time dynamic environmental prediction model of the next moment.
As shown in fig. 2, a shopping mall is exemplified, in order to improve customer experience, a mobile charging robot is introduced for customers to charge at any time when needed, however, the internal environment of the shopping mall is complex, and there are many obstacles such as pedestrians, commodity display shelves and elevators, so that the mobile charging robot needs to have high-efficiency obstacle avoidance capability, when the mobile charging robot operates, environment information in the shopping mall, including the position, shape, size and dynamic information (such as the moving speed of pedestrians) of the obstacles, is collected in real time, and is integrated into first environment data, the obstacle avoidance difficulty faced by the people is evaluated according to an algorithm according to the first environment data, for example, if fewer surrounding obstacles are detected, and the moving speed is slow, the robot determines the obstacle avoidance grade as primary, if the obstacles are more, but the moving rule can circulate, such as fixed commodity display shelves and pedestrians which move slowly, the robot determines the obstacle avoidance grade as intermediate grade, and if the environment is extremely complex, such as the pedestrian suddenly moves fast, and the robot determines the obstacle avoidance grade as high grade.
The mobile charging robot can detect surrounding obstacles by adopting a simple radio wave detection method, such as infrared rays or ultrasonic waves, when the obstacle avoidance level is primary, the robot can immediately adjust the driving direction of the robot to carry out rough obstacle avoidance once the obstacles are detected, the transmitting device of radio waves can be obtained by improving the transmitting end of the wireless charging device of the robot, when the obstacle avoidance level is intermediate, the robot can start a multi-sensor fusion system, second environment data acquired by sensors such as a laser radar and a camera are comprehensively used for predicting the movement track of the obstacles, for example, the robot can predict the future movement path of the robot by analyzing the walking speed and the direction of pedestrians and plan an optimal obstacle avoidance path according to the future movement path, when the obstacle avoidance level is high-level, the mobile charging robot can start an unmanned plane carried by the robot to carry out auxiliary scanning, the unmanned plane can be carried with a high-resolution camera and a sensor, the environment in a shopping center can be scanned in real time, the acquired third environment data can be returned to the robot, and the robot can be combined with the second environment data and the third environment data, the robot can predict the dynamic environment data, and the optimal obstacle avoidance path can be predicted by combining the real-time environment data and the dynamic environment data, and the dynamic obstacle avoidance model can be predicted, and the optimal obstacle avoidance path can be predicted.
Optionally, the determining the obstacle avoidance level according to the first environmental data includes:
judging whether an obstacle exists around according to the first environmental data;
if no obstacle exists around, the obstacle avoidance function is not started;
if the surrounding obstacle exists, judging the obstacle avoidance grade.
The mobile charging robot is provided with various sensors, such as a camera, a laser radar, a millimeter wave radar and the like, and can acquire surrounding environment information in real time, including the number, the shape, the distance, the speed and the like of objects, the information is summarized into first environment data, the robot firstly judges whether obvious obstacles exist around according to the first environment data, if the environment is wide, the robot does not start an obstacle avoidance function and continues to run along a preset path without any object obstructing the progress of the environment, and if the surrounding obstacles are detected, the robot further judges the obstacle avoidance grade according to the factors such as the type, the number, the distance, the moving speed and the like of the obstacles.
Optionally, if there is an obstacle around, determining the obstacle avoidance level further includes:
Judging the state of the obstacle according to the first environmental data;
If the obstacle is in a static state, judging the obstacle avoidance grade as a primary grade;
If the obstacle is in a dynamic state, judging the obstacle avoidance grade according to the number of times, the number and the speed of direction change of the obstacle.
The mobile charging robot starts working in an exemplary running process, acquires environmental information in a shopping center, such as the position, the shape, the size and dynamic information (such as the moving speed of pedestrians) of an obstacle in real time, integrates the information into first environmental data to provide basis for subsequent obstacle avoidance decisions, judges whether the surrounding obstacle exists according to the first environmental data, for example, the robot detects a static commodity display rack in front, the robot continuously judges the state of the obstacle due to the existence of the obstacle, the static commodity display rack is judged to be a static obstacle, the obstacle avoidance grade corresponding to the static obstacle is set as the primary, and if the robot detects the moving pedestrian in front, the robot judges the moving pedestrian to be a dynamic obstacle, the robot needs to evaluate the direction change times, the number and the speed of the dynamic obstacle (the pedestrian) so as to judge the obstacle avoidance grade.
Optionally, the step of judging the obstacle avoidance level according to the number, the number and the speed of the direction change of the obstacle includes:
setting a first weight factor for the number of direction changes of the obstacle Setting a second weight factor for the speed of the obstacle;
Specifically, a first weight factorAnd a second weight factorThe importance of the number of direction changes and the speed in judging the obstacle avoidance level is reflected.
Scoring the number, quantity and speed of direction changes of the obstacle, and constructing a scoring matrix of the number and speed of direction changes of the obstacleWherein each row represents an obstacle, each column represents an evaluation condition (including the number of direction changes and the speed), and the scoring matrixThe structure of (2) is as follows:
,
in the formula, Is the firstA direction change number score value for each obstacle; Is the first A speed score value for each obstacle;
Specifically, the number and speed of direction changes of each obstacle are scored, and a scoring matrix is constructed Scoring matrixEach row of (a) represents an obstacle, each column represents an evaluation condition (including the number of direction changesSum speed of)。
Calculating the direction change number weighted score and the speed weighted score of each obstacle, and adding the direction change number weighted score and the speed weighted score of each obstacle to obtain a weighted total score of each obstacle, wherein the weighted total score of any one obstacleThere is
;
Adding the weighted total score value of all the barriers to the score value of the number of barriers to obtain an actual score valueFor the actual scoring valueHas the following components
,
In the formula,Is the number of obstacles; Scoring the number of obstacles;
and judging the obstacle avoidance grade according to the actual grading value and the first threshold value.
Optionally, scoring the number, number and speed of direction changes of the obstacle includes:
Setting a second threshold of the number of times of changing the direction of the obstacle, setting a third threshold of the speed of the obstacle, and setting a fourth threshold of the number of the obstacles;
if the number of the direction changes of the obstacle is smaller than the second threshold value, obtaining a first score, otherwise, obtaining a second score;
If the speed of the obstacle is smaller than the third threshold value, a third score is obtained, otherwise, a fourth score is obtained;
And if the number of the barriers is smaller than the fourth threshold value, obtaining a fifth score, otherwise, obtaining a sixth score.
Optionally, the determining the obstacle avoidance grade according to the actual grading value and the first threshold value includes:
Judging the actual grading value and the first threshold value;
If the actual grading value is smaller than the first threshold value, judging that the obstacle avoidance grade is a middle grade;
And if the actual grading value is greater than or equal to the first threshold value, judging that the obstacle avoidance grade is high.
Specifically, according to actual scoring valuesAnd a preset first threshold value, judging the obstacle avoidance grade, ifIf the obstacle avoidance grade is smaller than or equal to the first threshold value, judging the obstacle avoidance grade as the primary grade, ifAnd if the obstacle avoidance grade is greater than or equal to the first threshold value, judging that the obstacle avoidance grade is high.
Illustratively, a first weighting factor is set for the number of direction changes of the obstacleSetting a second weight factor for the speed of the obstacleAccording to the actual situation, a scoring standard is set for the direction change times and the speed of the obstacle, for example, the fewer the direction change times, the lower the score, the slower the speed, the lower the score, for example, the robot detects 3 obstacles (pedestrians) at a certain time point, respectively marked as obstacle 1, obstacle 2 and obstacle 3, a first threshold value is 7, a second threshold value is 1, a third threshold value is 2, a fourth threshold value is 4, a first score value, a third score value and a fourth score value are set to 1, a second score value, a fourth score value and a sixth score value are set to 2, the direction change times of the obstacle 1 are 0 times, the speed is 3, the direction change times of the obstacle 2 are 0 times, the speed is 1, the direction change times of the obstacle 3 are 0 times, the speed is 4, the direction change times and the speed of each obstacle are scored, and the direction change times and the speed of the obstacle 1 are obtainedObstacle 2Obstacle 3Constructing a scoring matrixCalculating a direction change number weighted score value and a speed weighted score value of each obstacle, the weighted score value of the obstacle 1 beingThe weighted score value of the obstacle 2 isThe weighted score value of the obstacle 3 isThereby calculating the actual score valueActual score value at this timeAnd if the obstacle avoidance grade is smaller than the first threshold 7, judging the obstacle avoidance grade as a middle grade.
Optionally, predicting the movement track of the obstacle according to the second environmental data, and planning the obstacle avoidance path to avoid the obstacle includes:
Establishing an environment sensing function, obtaining a first environment sensing result according to the second environment data and the environment sensing function, and regarding the first environment sensing result There is
,
In the formula,As a function of the perception of the environment,Is thatThe second environmental data is collected at the moment;
Illustratively, first, a context awareness function is defined The function is capable of receivingSecond environmental data collected at momentAnd outputs a first environment sensing resultEnvironment awareness functionAnd extracting information related to the position, the speed, the acceleration and the like of the obstacle through processing and analyzing the second environment data, so as to form real-time perception of the environment.
Establishing a motion prediction function, collecting historical motion data of an obstacle, combining the environmental perception result, obtaining a first obstacle motion trail prediction result through the motion prediction function, and obtaining the first obstacle motion trail prediction resultThere is
,
In the formula,Representation ofA first obstacle movement trajectory prediction result of the moment,The motion prediction function is represented by a function of motion,Representation ofObstacle historical motion data at time;
Illustratively, historical motion data of the obstacle needs to be collected before predicting the obstacle motion profile The data includes the position, speed, acceleration and other information of obstacle, the method for collecting historical motion data can be realized by continuous recording and analysis of sensor data, and a motion prediction function is definedThe function is capable of receiving context awareness resultsAnd obstacle historical motion dataAnd outputPrediction result of first obstacle movement track at momentMotion prediction functionAnd predicting the motion trail of the obstacle at the future moment by comprehensively considering the environment sensing result and the obstacle historical motion data and utilizing a mathematical model or a machine learning algorithm.
Establishing a first path planning function, obtaining a first path planning result according to the obstacle movement track prediction result and the first path planning function, and planning the first path planning resultThere is
,
In the formula,The result of the first path planning is indicated,Representing a first path planning function;
Illustratively, a first path planning function is defined The function is capable of receiving a motion prediction functionOutput result of (2)And outputs the first path planning resultPath planning functionAn optimal path from the current position to the target position is found by an algorithm while avoiding collisions with predicted obstacle trajectories.
And planning a robot motion path according to the first path planning result to realize obstacle avoidance.
Illustratively, the first path planning result is obtainedAnd the mobile charging robot can adjust the motion path of the mobile charging robot according to the result, and the mobile charging robot can drive along the planned path, and continuously monitor the change of the surrounding environment at the same time so as to re-plan the path when necessary.
In an exemplary embodiment, in the obstacle avoidance system of the mobile charging robot, when the obstacle avoidance level is determined to be a middle level, the robot starts a multi-sensor fusion system, predicts a motion track of an obstacle by using second environmental data acquired in real time by sensors such as a laser radar and a camera, and plans an optimal obstacle avoidance path according to the motion track.
Optionally, the constructing a real-time dynamic environment prediction model of the next moment according to the second environmental data and the third environmental data, and dynamically planning the obstacle avoidance path after the next moment by combining the real-time dynamic environment prediction model of the next moment includes:
calculating the environmental perception related parameters at the next moment according to the environmental perception function, the second environmental data and the third environmental data to obtain a second environmental perception result ;
Calculating the obstacle motion trail after the next moment according to the second environment sensing result, the motion prediction function and the first obstacle motion trail prediction result to obtain a second obstacle motion trail prediction result, and regarding the second obstacle motion trail prediction resultThere is
,
In the formula,Representation ofA second obstacle movement track prediction result at the moment;
According to the second environment sensing result and the second obstacle movement track prediction result, a real-time dynamic environment prediction model of the next moment is established, and for the real-time dynamic environment prediction model of the next moment There is
,
In the formula,Representing a function for constructing a real-time dynamic environment prediction model at the next moment;
Acquiring obstacle position coordinate information according to the third environment data;
Establishing a second path planning function, calculating a planned path after the next moment to obtain a second path planning result according to the second path planning function, the obstacle position coordinate information, the real-time dynamic environment model and the second obstacle movement track prediction result, and calculating the second path planning result according to the second path planning result There is
,
In the formula,A function is planned for the second path and,Is obstacle position coordinate information.
Illustratively, the environmental state information of the next moment is predicted according to the environmental perception function, the second environmental data and the third environmental data to obtain a second environmental perception resultThe robot uses the second context awareness resultAnd previous obstacle movement trajectory prediction resultsAs input by a motion prediction functionTo calculate the predicted result of the movement track of the obstacle after the next momentThe robot perceives the second environment as a resultAnd obstacle movement track prediction resultAs input, by constructing a function of the real-time dynamic environmental prediction model at the next momentTo construct a real-time dynamic environment prediction model at the next momentThe robot acquires position coordinate information of the obstacle from the third environment dataThis information is critical for path planning, and the robot builds a second path planning functionAnd according to the real-time dynamic environment prediction modelObstacle position coordinate informationObstacle movement trajectory prediction resultCalculating to obtain a second path planning resultThis path planning resultIs the obstacle avoidance path that the robot should follow after the next moment.
Optionally, the calculating the environmental perception related parameter at the next moment according to the environmental perception function, the second environmental data and the third environmental data to obtain a second environmental perception result includes:
performing target matching and space-time alignment on the second environmental data and the third environmental data to obtain fourth environmental data;
illustratively, the robot first performs object matching and space-time alignment on the second environmental data (such as the position, speed, etc. of the obstacle) and the known third environmental data (such as the position of the obstacle, road information, etc. on the map) acquired by the robot to ensure consistency and accuracy of the data, thereby obtaining fourth environmental data.
Establishing a prediction model for describing the change of the fourth environmental data with time;
Illustratively, the robot builds a predictive model describing the time-dependent change of the fourth environmental data, which may be machine-learning based, such as a time-series analysis model, a neural network model, or the like.
According to the fourth environmental data at the previous moment, calculating a predicted value of the fourth environmental data at the current moment through the prediction model;
Training the prediction model according to the fourth environmental data and the predicted value at the current moment to obtain a stable prediction model;
Calculating the fourth environmental data at the next moment according to the stable prediction model prediction and the fourth environmental data to obtain fifth environmental data;
The robot calculates a predicted value of the current moment through a prediction model by using fourth environmental data of the previous moment, and compares the predicted value with the fourth environmental data of the current moment which is actually collected to train and optimize the prediction model, and once the prediction model is stabilized, the robot can use the predicted value to predict the fourth environmental data of the next moment, which is called fifth environmental data.
Calculating environmental perception related parameters at the next moment according to the environmental perception function and the fifth environmental data to obtain a second environmental perception result, and regarding the second environmental perception resultThere are;
,
in the formula, Is the fifth environmental data.
Illustratively, the robot perceives the function according to the environmentAnd calculating the environmental perception related parameters of the next moment by the fifth environmental data to obtain a second environmental perception result =And predicting the motion trail after the next moment by using the environmental information data of the predicted next moment, so that the pre-judging control in advance is realized, and the obstacle avoidance efficiency of the robot is improved.
Illustratively, as shown in fig. 2, the robot collects information of surrounding environment in real time through sensors (such as an infrared sensor, an ultrasonic sensor, a laser radar and the like) equipped with the robot, including the position, the size, the shape, the movement state and the like of an obstacle, if the environment information collected by the robot indicates that the obstacle is not in front or the obstacle is far away, the robot will not start the obstacle avoidance function and continue to run along a preset path, the robot determines whether the obstacle avoidance function needs to be started by judging whether the collected environment information contains the obstacle, and if the obstacle is detected, the robot also needs to judge whether the obstacle is static or dynamic. The method comprises the steps of describing that a static obstacle has small influence on the running of a robot, describing that the obstacle has small influence on the running of the robot, adopting a simple radio wave detection method, planning a bypass path by the robot, planning the obstacle avoidance path by the robot according to the motion state of the obstacle for a dynamic obstacle, calculating an actual grading value by the robot according to factors such as the type, the size and the distance of the obstacle and the running speed of the robot, measuring the influence degree of the obstacle on the running of the robot, comparing the actual grading value with a preset first threshold by the robot, describing that the influence of the obstacle on the running of the robot is large if the actual grading value is smaller than the first threshold, starting a multi-sensor fusion system by the robot in the middle obstacle avoidance stage, predicting the motion track of the obstacle by using second environment data acquired by sensors such as a laser radar and a camera, and the like, describing that the influence of the obstacle on the running of the robot is very large by the obstacle is detected by the robot if the actual grading value is larger than or equal to the preset first threshold, predicting the motion track of the robot by using the robot, and carrying data of the robot in advance, and realizing the prediction of the obstacle at the next stage by the high moment by combining with the predicted motion data of the robot, and realizing the prediction of the unmanned obstacle.
Based on the same inventive concept, as shown in fig. 3, the invention further provides a mobile charging robot obstacle avoidance system based on multi-sensor fusion, wherein the system comprises:
The environment data acquisition module is used for acquiring surrounding environment information in real time to obtain first environment data;
The obstacle avoidance grade decision module is used for judging an obstacle avoidance grade according to the first environmental data;
The obstacle avoidance strategy execution module is used for detecting surrounding obstacles through the radio wave emission module if the obstacle avoidance grade is primary, acquiring second environmental data through a plurality of sensors and predicting an obstacle movement track according to the second environmental data to plan an obstacle avoidance path if the obstacle avoidance grade is secondary, and acquiring third environmental data through real-time scanning of an unmanned aerial vehicle if the obstacle avoidance grade is advanced, constructing a real-time dynamic environmental prediction model according to the second environmental data and the third environmental data, and dynamically planning the obstacle avoidance path by combining the real-time dynamic environmental prediction model;
And the motion control module is used for controlling the robot to move according to the control.
The electrical connection between the above units does not necessarily mean connection between lines, and an indirect connection method may be applied to the embodiments of the present invention as long as the object of the present invention is achieved. The foregoing is merely exemplary embodiments of the present invention and is not intended to limit the scope of the present invention.
That is, equivalent changes and modifications are contemplated by the teachings of the present application, which fall within the scope of the present application. Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains.
Claims (7)
1. The mobile charging robot obstacle avoidance method based on multi-sensor fusion is characterized by comprising the following steps of:
collecting surrounding environment information in real time to obtain first environment data;
judging an obstacle avoidance grade according to the first environmental data;
if the obstacle avoidance grade is primary, detecting surrounding obstacles by transmitting radio waves to perform rough obstacle avoidance;
If the obstacle avoidance grade is a middle grade, acquiring second environmental data through a plurality of sensors, predicting an obstacle movement track according to the second environmental data, and planning an obstacle avoidance path to avoid an obstacle;
If the obstacle avoidance grade is advanced, third environmental data is obtained through real-time scanning of the unmanned aerial vehicle, a real-time dynamic environment prediction model of the next moment is built according to the second environmental data and the third environmental data, and an obstacle avoidance path after the next moment is dynamically planned to avoid the obstacle by combining the real-time dynamic environment prediction model of the next moment;
Wherein, the judging the obstacle avoidance grade according to the first environmental data includes:
judging whether an obstacle exists around according to the first environmental data;
if no obstacle exists around, the obstacle avoidance function is not started;
if an obstacle exists around, judging the state of the obstacle according to the first environmental data;
If the obstacle is in a static state, judging the obstacle avoidance grade as a primary grade;
If the obstacle is in a dynamic state, judging the obstacle avoidance grade according to the number of times, the number and the speed of direction change of the obstacle;
setting a first weight factor for the number of direction changes of the obstacle Setting a second weight factor for the speed of the obstacle;
Scoring the number, quantity and speed of direction changes of the obstacle, and constructing a scoring matrix of the number and speed of direction changes of the obstacleWherein each row represents an obstacle, each column represents an evaluation condition, and a scoring matrixThe structure of (2) is as follows:
,
in the formula, Is the firstA direction change number score value for each obstacle; Is the first A speed score value for each obstacle;
Calculating the direction change number weighted score and the speed weighted score of each obstacle, and adding the direction change number weighted score and the speed weighted score of each obstacle to obtain a weighted total score of each obstacle, wherein the weighted total score of any one obstacle There is
;
Adding the weighted total score value of all the barriers to the score value of the number of barriers to obtain an actual score valueFor the actual scoring valueHas the following components
,
In the formula,Is the number of obstacles; Scoring the number of obstacles;
and judging the obstacle avoidance grade according to the actual grading value and the first threshold value.
2. The mobile charging robot obstacle avoidance method based on multi-sensor fusion as claimed in claim 1, wherein scoring the number, number and speed of direction changes of the obstacle comprises:
Setting a second threshold of the number of times of changing the direction of the obstacle, setting a third threshold of the speed of the obstacle, and setting a fourth threshold of the number of the obstacles;
if the number of the direction changes of the obstacle is smaller than the second threshold value, obtaining a first score, otherwise, obtaining a second score;
If the speed of the obstacle is smaller than the third threshold value, a third score is obtained, otherwise, a fourth score is obtained;
And if the number of the barriers is smaller than the fourth threshold value, obtaining a fifth score, otherwise, obtaining a sixth score.
3. The method for avoiding the obstacle of the mobile charging robot based on the multi-sensor fusion according to claim 2, wherein the determining the level of the obstacle avoidance according to the actual grading value and the first threshold value comprises:
Judging the actual grading value and the first threshold value;
If the actual grading value is smaller than the first threshold value, judging that the obstacle avoidance grade is a middle grade;
And if the actual grading value is greater than or equal to the first threshold value, judging that the obstacle avoidance grade is high.
4. The method for avoiding the obstacle of the mobile charging robot based on the multi-sensor fusion according to claim 1, wherein predicting the movement track of the obstacle according to the second environmental data, planning the obstacle avoidance path for avoiding the obstacle comprises:
Establishing an environment sensing function, obtaining a first environment sensing result according to the second environment data and the environment sensing function, and regarding the first environment sensing result There is
,
In the formula,As a function of the perception of the environment,Is thatThe second environmental data is collected at the moment;
establishing a motion prediction function, collecting historical motion data of an obstacle, combining the environmental perception result, obtaining a first obstacle motion trail prediction result through the motion prediction function, and obtaining the first obstacle motion trail prediction result There is
,
In the formula,Representation ofA first obstacle movement trajectory prediction result of the moment,The motion prediction function is represented by a function of motion,Representation ofObstacle historical motion data at time;
establishing a first path planning function, obtaining a first path planning result according to the obstacle movement track prediction result and the first path planning function, and planning the first path planning result There is
,
In the formula,The result of the first path planning is indicated,Representing a first path planning function;
And planning a robot motion path according to the first path planning result to realize obstacle avoidance.
5. The method for avoiding the obstacle of the mobile charging robot based on the multi-sensor fusion according to claim 4, wherein the constructing a real-time dynamic environment prediction model of the next moment according to the second environment data and the third environment data, and dynamically planning an obstacle avoidance path after the next moment by combining the real-time dynamic environment prediction model of the next moment comprises:
calculating the environmental perception related parameters at the next moment according to the environmental perception function, the second environmental data and the third environmental data to obtain a second environmental perception result ;
Calculating the obstacle motion trail after the next moment according to the second environment sensing result, the motion prediction function and the first obstacle motion trail prediction result to obtain a second obstacle motion trail prediction result, and regarding the second obstacle motion trail prediction resultThere is
,
In the formula,Representation ofA second obstacle movement track prediction result at the moment;
According to the second environment sensing result and the second obstacle movement track prediction result, a real-time dynamic environment prediction model of the next moment is established, and for the real-time dynamic environment prediction model of the next moment There is
,
In the formula,Representing a function for constructing a real-time dynamic environment prediction model at the next moment;
Acquiring obstacle position coordinate information according to the third environment data;
Establishing a second path planning function, calculating a planned path after the next moment to obtain a second path planning result according to the second path planning function, the obstacle position coordinate information, the real-time dynamic environment prediction model and the second obstacle movement track prediction result, and calculating the second path planning result for the second path planning result There is
,
In the formula,A function is planned for the second path and,Is obstacle position coordinate information.
6. The method for avoiding an obstacle for a mobile charging robot based on multi-sensor fusion according to claim 5, wherein calculating the environmental perception related parameters at the next moment according to the environmental perception function, the second environmental data and the third environmental data to obtain a second environmental perception result comprises:
performing target matching and space-time alignment on the second environmental data and the third environmental data to obtain fourth environmental data;
establishing a prediction model for describing the change of the fourth environmental data with time;
According to the fourth environmental data at the previous moment, calculating a predicted value of the fourth environmental data at the current moment through the prediction model;
Training the prediction model according to the fourth environmental data and the predicted value at the current moment to obtain a stable prediction model;
Calculating the fourth environmental data at the next moment according to the stable prediction model prediction and the fourth environmental data to obtain fifth environmental data;
calculating environmental perception related parameters at the next moment according to the environmental perception function and the fifth environmental data to obtain a second environmental perception result, and regarding the second environmental perception result There are;
,
in the formula, Is the fifth environmental data.
7. A mobile charging robot obstacle avoidance system based on multi-sensor fusion, applied to the obstacle avoidance method of any one of claims 1-6, characterized in that the system comprises:
The environment data acquisition module is used for acquiring surrounding environment information in real time to obtain first environment data;
The obstacle avoidance grade decision module is used for judging an obstacle avoidance grade according to the first environmental data;
The obstacle avoidance grade decision module is further used for judging whether an obstacle exists around according to the first environmental data, if the obstacle does not exist around, the obstacle avoidance function is not started, if the obstacle exists around, the state of the obstacle is judged according to the first environmental data, and if the obstacle is static, the obstacle avoidance grade is judged to be primary;
the obstacle avoidance grade decision module is also used for calculating an actual grading value according to the direction change times, the number and the speed of the obstacle if the obstacle is in a dynamic state, and judging the obstacle avoidance grade to be medium grade or high grade according to the actual grading value and the first threshold value;
The obstacle avoidance strategy execution module is used for detecting surrounding obstacles through the radio wave emission module if the obstacle avoidance grade is primary, acquiring second environmental data through a plurality of sensors and predicting an obstacle movement track according to the second environmental data to plan an obstacle avoidance path if the obstacle avoidance grade is secondary, and acquiring third environmental data through real-time scanning of an unmanned aerial vehicle if the obstacle avoidance grade is advanced, constructing a real-time dynamic environmental prediction model according to the second environmental data and the third environmental data, and dynamically planning the obstacle avoidance path by combining the real-time dynamic environmental prediction model;
and the motion control module is used for controlling the robot to move according to the obstacle avoidance grade.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411625856.8A CN119148722B (en) | 2024-11-14 | 2024-11-14 | A mobile charging robot obstacle avoidance method and system based on multi-sensor fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411625856.8A CN119148722B (en) | 2024-11-14 | 2024-11-14 | A mobile charging robot obstacle avoidance method and system based on multi-sensor fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119148722A CN119148722A (en) | 2024-12-17 |
| CN119148722B true CN119148722B (en) | 2025-02-18 |
Family
ID=93814036
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411625856.8A Active CN119148722B (en) | 2024-11-14 | 2024-11-14 | A mobile charging robot obstacle avoidance method and system based on multi-sensor fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119148722B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120462388B (en) * | 2025-07-11 | 2025-11-07 | 东营鑫翔石油装备有限公司 | A derrick mounting bracket and a self-propelled control method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117970925A (en) * | 2023-12-29 | 2024-05-03 | 上海交大智邦科技有限公司 | Robot real-time obstacle avoidance and dynamic path planning method and system |
| CN118068836A (en) * | 2024-02-21 | 2024-05-24 | 南京邮电大学 | Multi-AGV obstacle avoidance and path planning method and system based on deep learning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112068553B (en) * | 2020-08-20 | 2024-08-20 | 上海姜歌机器人有限公司 | Robot obstacle avoidance processing method and device and robot |
| CN118426471A (en) * | 2024-04-26 | 2024-08-02 | 浙江师范大学 | Intelligent service robot used in complex environment |
| CN118938919A (en) * | 2024-07-24 | 2024-11-12 | 大连理工大学 | A dynamic coordination obstacle avoidance method for intelligent handling robots based on reinforcement learning |
-
2024
- 2024-11-14 CN CN202411625856.8A patent/CN119148722B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117970925A (en) * | 2023-12-29 | 2024-05-03 | 上海交大智邦科技有限公司 | Robot real-time obstacle avoidance and dynamic path planning method and system |
| CN118068836A (en) * | 2024-02-21 | 2024-05-24 | 南京邮电大学 | Multi-AGV obstacle avoidance and path planning method and system based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119148722A (en) | 2024-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ferrer et al. | Bayesian human motion intentionality prediction in urban environments | |
| Chen et al. | Robot navigation with map-based deep reinforcement learning | |
| KR20220055167A (en) | Autonomous robot, world map management server of autonomous robot and collision avoidance method using the same | |
| CN119148722B (en) | A mobile charging robot obstacle avoidance method and system based on multi-sensor fusion | |
| CN119190018B (en) | A V2X-based blind spot assisted driving method and system for intersections | |
| CN119919821B (en) | Safety analysis method and system applied to storage robot complex driving scene | |
| CN119283848A (en) | A control method for unmanned vehicles based on deep learning | |
| CN119691513A (en) | Dynamic risk assessment method combining visual positioning and mapping with control barrier function | |
| Kim et al. | Adaptive sensor management for UGV monitoring based on risk maps | |
| Wang et al. | The application of computer vision target recognition technology in autonomous driving | |
| CN118605538B (en) | Multi-source-based target data collaborative processing method and system | |
| Rezaei et al. | A deep learning-based approach for vehicle motion prediction in autonomous driving | |
| CN120190815A (en) | An AI-driven robot working environment safety perception method | |
| CN120428724A (en) | A visual acquisition and avoidance control method for AGV robots based on deep thinking | |
| CN120010496A (en) | Intelligent luggage obstacle avoidance method and system based on wireless sensor | |
| Yang | Development of obstacle avoidance for autonomous vehicles and an optimization scheme for the artificial potential field method | |
| Kothuri et al. | Enhancing Pedestrian Safety in Autonomous Vehicles through Machine Learning | |
| CN117429419B (en) | Automatic driving method applied to port and driving vehicle | |
| CN120972982B (en) | AGV automatic driving control method and controller based on positioning sensing | |
| Hongsheng | Far-field sensing in partial VANET environment | |
| CN120702480B (en) | A Smart Navigation Method for Robots | |
| Jha et al. | Watch Out for the Safety-Threatening Actors: Proactively Mitigating Safety Hazards | |
| Shanmugam et al. | Integrating Advanced Machine Learning, Sensor Fusion, and Control Systems to Enhance Autonomous Vehicle Safety and Performance | |
| Particke et al. | Entropy-based intention change detection with a multi-hypotheses filter | |
| Kumar et al. | Object Detection using Machine Learning to Avoid Accident |
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 |