CN107891423B - Intelligent exploration robot based on multi-sensor fusion detection and detection method thereof - Google Patents
Intelligent exploration robot based on multi-sensor fusion detection and detection method thereof Download PDFInfo
- Publication number
- CN107891423B CN107891423B CN201711089041.2A CN201711089041A CN107891423B CN 107891423 B CN107891423 B CN 107891423B CN 201711089041 A CN201711089041 A CN 201711089041A CN 107891423 B CN107891423 B CN 107891423B
- Authority
- CN
- China
- Prior art keywords
- support
- sensor
- singlechip
- wheel
- detection method
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 51
- 230000004927 fusion Effects 0.000 title claims abstract description 31
- 238000013016 damping Methods 0.000 claims abstract description 32
- 230000007246 mechanism Effects 0.000 claims abstract description 26
- 230000009471 action Effects 0.000 claims abstract description 10
- 230000003993 interaction Effects 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 24
- 230000001360 synchronised effect Effects 0.000 claims description 24
- 230000005540 biological transmission Effects 0.000 claims description 23
- 238000000034 method Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 11
- 238000007499 fusion processing Methods 0.000 claims description 10
- 230000007613 environmental effect Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000012706 support-vector machine Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims description 5
- 238000004064 recycling Methods 0.000 claims description 4
- 229920000049 Carbon (fiber) Polymers 0.000 claims description 3
- 239000004917 carbon fiber Substances 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 239000007789 gas Substances 0.000 description 9
- 238000010276 construction Methods 0.000 description 6
- 238000011156 evaluation Methods 0.000 description 5
- 230000033001 locomotion Effects 0.000 description 5
- 230000035939 shock Effects 0.000 description 4
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 241000239290 Araneae Species 0.000 description 1
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 241000238631 Hexapoda Species 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 229910002091 carbon monoxide Inorganic materials 0.000 description 1
- 239000003638 chemical reducing agent Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000011664 nicotinic acid Substances 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000007665 sagging Methods 0.000 description 1
- 229910001220 stainless steel Inorganic materials 0.000 description 1
- 239000010935 stainless steel Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1628—Programme controls characterised by the control loop
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J13/00—Controls for manipulators
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
Abstract
The invention relates to an intelligent exploration robot based on multi-sensor fusion detection and a detection method thereof, wherein the intelligent exploration robot comprises a support, a crawler chassis and a damping mechanism are respectively arranged on two sides of the support, and the damping mechanism is positioned on the crawler chassis; the thermistors, the flame sensors, the ultrasonic sensors and the infrared sensors are arranged on two sides of the support, and the thermistors are evenly distributed on two sides of the support; the infrared sensor and the ultrasonic sensor are arranged at the front non-shielding position of the support; the flame sensor is arranged on the support and can contact with a larger and more open position of the air area; the inside of the support is provided with a singlechip, a gear motor, a signal receiving station and a collision sensor, wherein the singlechip is used for receiving signals transmitted by the collision sensor, the thermistor, the flame sensor, the ultrasonic sensor and the infrared sensor and controlling the action of the gear motor; the signal station performs information interaction with the singlechip; a mechanical arm is arranged at the top of the support, and the mechanical arm is controlled to work by a singlechip; a power supply is also arranged inside the support.
Description
Technical Field
The invention relates to an exploration robot and a detection method thereof, in particular to an intelligent exploration robot based on multi-sensor fusion detection and a detection method thereof.
Background
With the continuous development of society, the human living standard is advancing, the safety of the robot and the exploration of the unknown field are also more and more important, more and more people begin to consider using robots to replace people to perform dangerous area work, such as tunnel construction, earthquake rescue, exploration of the unknown area, military detection, even planet detection and the like, the work is often not dry-tied with complex and dangerous environment, if the performance of the robot is not strong enough, the detection is not accurate enough, the difficult work cannot be completed when the action range is not enough, all of the requirements are that the robot has high environment detection and adaptability, and the adaptability to various terrains and the accurate dangerous detection capability, otherwise, the robot cannot replace the people to complete the task.
When detecting a dangerous area or rescuing a disaster area, the detection of dangerous signals is particularly important, and from common light rays, sound waves, detection of dangerous gases (carbon monoxide, methane, nitrogen and the like), measurement of temperature at a fire place and the like, the data can often help people to know disaster situations or construction information, and the situation of the scene has a global grasp, so that rescue strategies and construction directions are determined. However, a large number of people enter a dangerous area to often cause secondary disaster and secondary casualties, so that the robot can replace human beings to complete detection tasks under the condition of the shortest time and the highest accuracy. In engineering, for example, when an operator works a tunnel, the detection robot can enter a construction site before the operator enters the construction site, and detect dangerous factors such as temperature, gas and the like in advance, so that the safety of the constructor is ensured, which is necessary for a plurality of large-scale construction projects.
At present, the detecting robot is a very hot topic worldwide, many countries are engaged in research on the aspect, the best-known is Boston dynamics (Boston dynamics) which are researched and developed to be the robots of the series such as Spotmini and Atlas, and the like, have very outstanding performance on the movement mode and the action stability, and are also very easy to deal with when the specific tasks are completed by virtue of high-precision sensors. In China, the detection robot also has remarkable breakthrough, and a plurality of categories exist in the movement mode. For example, a wheeled detection robot relies on ultrasonic, infrared and other sensors to detect the environment, however, the fact that the wheel type is not suitable for complex terrains is a fatal disadvantage, while a crawler type detection robot has the unusual capability of detecting the environment and adapting to terrains, but the slow speed is the biggest obstacle for completing the detection task. In order to solve the problems, some companies in China put forward bionic robots, such as snake-shaped machines, which have the advantages of high movement speed and small friction, but are suitable for specific environments and have single functions; hexapod robots are similar to a spider and have a strong adaptability to terrain, but their speed on flat ground and response speed during operation are not as good as wheeled robots.
Functionally, the existing detection robots mainly detect environments, detect dangers and feed back information, but most of the robots on the market at present mainly detect single sensors, and rarely integrate multiple sensors to perform comprehensive judgment, so that the anti-interference capability and the precision capability of the robots are poor, the robots cannot adapt to severe environments, and the working range of the robots is mostly in places with stable terrains and low personnel concentration.
Disclosure of Invention
Aiming at the problems, the invention aims to provide an intelligent exploration robot based on multi-sensor fusion detection and a detection method thereof, which can adapt to severe environments and realize dead angle-free detection in complex variable working condition environments.
In order to achieve the above purpose, the present invention adopts the following technical scheme: intelligent exploration robot based on multisensor fuses detection, its characterized in that: the robot comprises a support, wherein a crawler chassis and a damping mechanism are respectively arranged on two sides of the support, and the damping mechanism is positioned on the crawler chassis; the two sides of the support are also provided with thermistors, flame sensors, ultrasonic sensors and infrared sensors, and the thermistors are evenly distributed on the two sides of the support; the infrared sensor and the ultrasonic sensor are arranged at the front non-shielding position of the support; the flame sensor is arranged at a position on the support, which can be in contact with air with larger and more open area; the single-chip microcomputer is used for receiving signals transmitted by the collision sensor, the thermistor, the flame sensor, the ultrasonic sensor and the infrared sensor and controlling the action of the speed-reducing motor; the signal station performs information interaction with the singlechip; the mechanical arm is arranged at the top of the support and is controlled by the singlechip to work; a power supply is also arranged in the support.
Further, a plurality of signal stations are arranged in the support, and each signal station is provided with a magnet device; the front end of the mechanical arm is provided with a micro electromagnet sucker controlled by the singlechip, and the micro electromagnet sucker is used for sucking the magnet device fixed on the signal station so as to place or take away the signal station.
Further, each crawler chassis comprises a guide wheel, a crawler belt, a riding wheel, a driving wheel, a transmission shaft, a thrust wheel, a driving shaft, a toothed belt and a synchronous belt wheel; an output shaft of the speed reducing motor is coaxially connected with the synchronous pulley, and the synchronous pulley is connected with the transmission shaft through the toothed belt; the transmission shaft is connected with the driving shaft through a synchronous pulley transmission mechanism, and the driving wheel is arranged on the driving shaft; the guide wheels, the riding wheels, the driving wheels and the supporting wheels are all arranged on the support and form a frame of the crawler chassis, and the crawler is positioned at the outer side of the frame and driven to move by the driving wheels; the guide wheel is positioned in front of the support, the driving wheel is positioned behind the support, and the height between the center position of the guide wheel and the ground is larger than that between the center position of the driving wheel and the ground; the supporting wheel is positioned between the guide wheel and the driving wheel, and the height between the central position of the supporting wheel and the ground is larger than that between the central position of the guide wheel and the ground; the thrust wheel is arranged on the damping mechanism.
Further, the synchronous pulley transmission mechanism comprises the synchronous pulley and a belt; the transmission shaft is coaxially provided with the synchronous pulleys, the driving shaft is also provided with the synchronous pulleys, and the two synchronous pulleys are in transmission connection through the belt.
Further, each of the shock absorbing mechanisms comprises a buffer spring and a shock absorbing arm; the two damping arms are arranged in a crossing way, the crossing points are movably connected to the side wall of the support, the two end parts of the damping arms above are arranged in a track in a sliding way, and the track is fixed on the side wall of the support; the two ends of the damping arm positioned below are respectively connected with the thrust wheel; and arc-shaped arms which are integrally formed with each damping arm are arranged above the crossing points, the two arc-shaped arms are correspondingly arranged to form a semicircular arc, and the end parts of the two arc-shaped arms are provided with the damping springs.
Further, the support is made of carbon fiber plates.
Further, a plurality of environment sensors are arranged on the support and are evenly distributed close to the top of the support; the laser ranging sensors are arranged at the edges of the two sides of the support and used for monitoring the positions of the robots, and data acquired by the environmental sensors and the laser ranging sensors are transmitted to the single chip microcomputer.
The intelligent exploration robot detection method based on multi-sensor fusion detection is characterized by comprising the following steps of: 1) Setting starting and ending positions; 2) The singlechip controls the gear motor to work and keeps the advancing state of the robot; 3) Each infrared sensor, each laser ranging sensor, each ultrasonic sensor and each environmental sensor perform data acquisition according to a preset rule, and the acquired data are transmitted into the singlechip; 4) Carrying out algorithm fusion processing according to the received data, obtaining an optimal path, feeding back to the singlechip, and returning to the step 2) to control the action of the gear motor so as to drive the driving wheel to advance in the optimal path; 5) Judging whether the optimal path is wrong or not, and if the optimal path is wrong, continuing to advance; if the optimal path is wrong, path adjustment is realized through an error adjustment mechanism, and then the action of the speed reduction motor is controlled by the singlechip; 6) Storing the obtained optimal path in real time and updating the optimal path in real time, so that the optimal path information is ensured to be used later after the exploration is finished; 7) Detecting and judging the current position, if the current position reaches the maximum wifi signal distance, placing a signal station, storing the current position, and detecting the current position again; otherwise, continuing to advance; 8) Returning according to the final path after reaching the end point, storing the final path, and recycling the signal station; 9) And evaluating the running path of the robot according to data transmitted by an environment sensor arranged on the robot support, and storing a corresponding evaluation result so as to be convenient for preliminary understanding of the working condition of the detected environment.
Further, if a fire disaster occurs, after the flame sensor on the support generates signals, the singlechip models the ambient temperature according to data transmitted by a plurality of thermistors around the support: according to the installation position of the thermistor and the transmitted data, surrounding space points are abstracted into a one-dimensional matrix containing x, y, z coordinates and temperature T, an isothermal line is established by finding out points with the same temperature, a normal direction unit vector of the isothermal line is calculated according to the coordinates of the points, and a calculation formula of a temperature gradient is obtained by the normal direction unit vector:
wherein n is a normal direction unit vector, and DeltaT is a temperature difference in the normal direction; Δn is the distance between the two isotherms corresponding to the temperature difference Δt in the normal direction thereof; and automatically advancing to the lowest temperature direction, namely the gradient minimum direction according to a calculation formula of the temperature gradient.
Further, in the step 4), the algorithm fusion processing procedure is as follows: 4.1 Classifying the received data; 4.2 The multi-information fusion algorithm of the multi-core function support vector machine is adopted, and a linear kernel function, a polynomial kernel function and a radial basis function method are mainly adopted to carry out support vector machine analysis on sound, force, temperature and light signals in the running process of the robot, obtain a global optimal solution, obtain an optimal kernel function and send the optimal kernel function to a fusion rule center to obtain a fusion rule set; wherein the fusion rule set is expressed as Γ= { r 1 ,…,r n -where r i (i=1, 2, …, n) represents the decision rule of the i-th sensor; 4.3 Obtaining an expected risk value according to the fusion rule set, and if the expected risk value is the minimum expected risk value R (Γ), entering the next step; otherwise, returning to the step 4.1); wherein the minimum expected risk value R (Γ) is:
wherein H is 1 Indicating the presence of a faulty target; h 0 Indicating that no fault target exists; l (L) ij As H j True time is judged as H i Probability of (2); p (P) j To assume H j Is (i, j=0, 1); 4.4 And (3) obtaining n groups of judgment rules through the minimum expected risk function solution, namely outputting a feature set, and completing multi-information fusion processing to select an optimal path.
Due to the adoption of the technical scheme, the invention has the following advantages: 1. according to the invention, a plurality of different types of sensors are adopted, and a multi-core function support vector machine fusion algorithm is utilized to optimize the advancing path of the robot, so that the robot can realize dead angle-free detection in a complex variable working condition environment. 2. The invention adopts closed-loop control, has an error adjustment mechanism for the travel path of the robot, and avoids the robot from entering into dead circulation. 3. The invention adopts the mode of the signal base station, thereby avoiding the situation that the lost signal is out of contact with the outside under special working conditions. 4. The invention outputs the exploration environment evaluation report while completing exploration, and provides dangerous values for users. 5. The steel track adopted by the invention has enough strength and rigidity, good wear resistance, good adhesion performance with the ground, and damping springs, and can isolate the influence of the terrain on the sensor (especially the collision sensor). Meanwhile, the robot is provided with a GPS module, so that the position information of the current robot can be fed back at any time.
Drawings
FIG. 1 is a schematic view of a robot configuration of the present invention;
FIG. 2 is a side view of the robot of the present invention;
FIG. 3 is a schematic flow chart of the detection method of the present invention;
fig. 4 is a schematic diagram of an algorithm fusion process of the present invention.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings and examples.
As shown in fig. 1 and 2, the present invention provides an intelligent exploration robot based on multi-sensor fusion detection, which is an improvement on the existing robots. The intelligent exploration robot comprises a support 1, wherein a crawler chassis and a damping mechanism 2 are respectively arranged on two sides of the support 1, and the damping mechanism 2 is located on the crawler chassis. The two sides of the support 1 are also provided with thermistors (namely temperature sensors), flame sensors, ultrasonic sensors and infrared sensors 3; the thermistors are arranged at positions far away from the gear motor 4 and the power supply 8 and are evenly distributed on two sides of the support 1; the infrared sensor 3 and the ultrasonic sensor are arranged at the front of the support at the position where no shielding exists; the flame sensor is arranged on the support 1 and can contact with a larger and more open position of the air area. The inside of the support 1 is provided with a singlechip, a gear motor 4, a signal station and a collision sensor 5, wherein the singlechip is used for receiving signals transmitted by the collision sensor 5, a thermistor, a flame sensor, an ultrasonic sensor and an infrared sensor 3 and controlling the gear motor 4 to act; and the signal station performs information interaction with the singlechip. The mechanical arm 6 is arranged at the top of the support 1, the specific structure of the mechanical arm 6 adopts the existing structure, and the details are not repeated here, and a rotating motor configured in the mechanical arm 6 is also controlled to work by a single chip microcomputer, so that the mechanical arm 6 is controlled to act. An antenna 7 is arranged on the top of the support 1 at the rear part of the mechanical arm 6, and the antenna 7 assists the signal station in receiving WiFi signals. Inside the holder 1 a power supply 8 is also provided for powering the whole robot.
In the above embodiments, the infrared sensor and the laser ranging sensor are arranged at intervals on the side surface of the support 1, and the ultrasonic sensor is arranged at the front part of the support 1. The infrared sensor and the laser ranging sensor can always act, and the ultrasonic sensor only acts on every other sensor at the same time, so that the two groups of sensors are alternately detected to avoid mutual interference.
In the above embodiment, the support 1 is further provided with a plurality of signal stations, and each signal station is provided with a magnet device. The front end of the mechanical arm 6 is provided with a micro electromagnet sucker 20 controlled by a single chip microcomputer, and a magnet device fixed on the signal station is sucked through the micro electromagnet sucker 20, so that the signal station is placed or taken away.
In the above embodiment, each crawler chassis includes the timing pulley 9, the toothed belt 10, the propeller shaft 11, the drive shaft 12, the drive wheel 13, the guide wheel 14, the carrier pulley 15, the bogie wheel 16, and the crawler 17. The output shaft of the gear motor 4 is coaxially connected with the synchronous pulley 9, and the synchronous pulley 9 is connected with the transmission shaft 11 through a toothed belt 10, so that the power of the gear motor 4 is transmitted to the transmission shaft 11. The transmission shaft 11 is connected with the driving shaft 12 through a synchronous pulley transmission mechanism, so as to drive the driving shaft 12 to rotate; the drive shaft 12 is provided with a drive wheel 13. The guide wheel 14, the riding wheel 15, the driving wheel 13 and the supporting wheel 16 are all arranged on the support 1 and form a frame of the crawler chassis, and the crawler 17 is positioned outside the frame and is driven to move by the driving wheel 13. The guide wheel 14 is positioned in front of the support 1 (the left side in fig. 2 is the front), the driving wheel 13 is positioned behind the support 1, the height from the ground of the center position of the guide wheel 14 is larger than the height from the ground of the center position of the driving wheel 13, and the driving wheel 13 is contacted with the ground through the crawler 17; the guide wheel 14 guides the crawler belt 17 to revolve correctly, so that the deviation and the derailment are prevented, and the lower part of the crawler belt is not easy to arch by the driving wheel 13. The riding wheel 15 is positioned between the guide wheel 14 and the driving wheel 13, the height from the center position of the riding wheel 15 to the ground is larger than the height from the center position of the guide wheel 14 to the ground, the riding wheel 15 drags the crawler 17, so that the crawler 17 is prevented from sagging too much, the jolt phenomenon of the crawler 17 in movement is reduced, and the crawler 17 is prevented from sliding sideways; the guide wheel 14, the riding wheel 15 and the driving wheel 13 form a triangle-like structure. The thrust wheel 16 is arranged on the damping mechanism 2 and is used for supporting the weight of the support 1 and distributing the weight on the crawler 17, so that the pressure distribution of the crawler 17 is uniform during grounding; and the caterpillar tracks 17 are not slipped laterally by means of the roller flanges of the thrust wheels 16.
In the above embodiment, the timing pulley transmission mechanism includes the timing pulley 9 and the belt; the transmission shaft 11 is coaxially provided with a synchronous pulley 9, the driving shaft 12 is also provided with a synchronous pulley 9, and the two synchronous pulleys 9 are in transmission connection through a belt, so that the power of the transmission shaft 11 is transmitted to the driving shaft 12.
In the above embodiment, the crawler belt 17 is a steel crawler belt having sufficient strength and rigidity and excellent wear resistance.
In the above embodiments, each damper mechanism 2 includes the damper arm 18 and the damper spring 19; two shock absorbing arms 18 are used. The two damping arms 18 are arranged in a crossing way, the crossing points are movably connected to the side wall of the support 1, the two end parts of the damping arm positioned above are arranged in the track in a sliding way, and the track is fixed on the side wall of the support 1; the two ends of the shock absorbing arm positioned below are respectively connected with the thrust wheel 16. An arc arm integrally formed with each damping arm 18 is further arranged above the intersection, the two arc arms are correspondingly arranged to form a semicircular arc, buffer springs 19 are arranged at the ends of the two arc arms to play a role in buffering and damping, and the buffer springs 19 are high-stiffness stainless steel springs. In use, the two damping arms 18 can rotate at the crossing point to adaptively deflect, are connected with the damping springs 19 to jointly play a damping role, and are matched with the thrust wheel 16 to adapt to the deformation of the terrain.
In the above embodiments, the support 1 is made of carbon fiber plate to reduce the weight. The support 1 is also provided with environmental sensors such as a harmful gas sensor, a temperature and humidity sensor and the like, and a plurality of environmental sensors are evenly distributed close to the top of the support 1; the laser ranging sensors used for monitoring the positions of the robot body are further arranged at the edges of the two sides of the support 1, and data acquired by the environment sensors and the laser ranging sensors are transmitted to the singlechip.
In the above embodiments, the signal station is used for remote data transmission, and the map 8266 module is used to generate wifi signal back data (location information, environment information, etc.). For the situation that the signal distance beyond the distance is not reached, a plurality of esp8266 modules are adopted as signal stations, and the mechanical arm can be used for placing and withdrawing the signal stations. And in the walking process, when the signal is at the maximum distance, feeding back the signal to the singlechip, stopping advancing, placing the signal station and recording the current position. And recycling the signal station according to the memory position after the search is finished.
In the above embodiments, the infrared sensor is GP2Y0a02YK0F, the laser ranging sensor is CHEM10-N, and the ultrasonic sensor is JS-SR04T.
In each embodiment, the model adopted by the singlechip is STM32F103ZE. Meanwhile, a GPS module connected with the singlechip is also arranged in the support 1, so that the position information of the current robot can be fed back at any time.
As shown in fig. 3, the invention further provides an intelligent exploration robot detection method based on multi-sensor fusion detection, which comprises the following steps:
1) Setting starting and ending positions, and starting automatic identification by a GPS module;
2) The singlechip controls the gear motor to work and keeps the advancing state of the robot;
3) Each infrared sensor, each laser ranging sensor, each ultrasonic sensor and each environmental sensor perform data acquisition according to a preset rule so as to avoid mutual interference, and the acquired data are transmitted into the singlechip;
wherein, the rule of presetting is: adjacent sensors on the support 1 do not operate simultaneously and receive data alternately at preset time intervals.
4) And (3) carrying out algorithm fusion processing according to the received data, obtaining an optimal path, feeding back to the singlechip, and returning to the step (2) to control the motion of the gear motor 4 so as to drive the driving wheel 13 to advance in the optimal path.
5) Judging whether the optimal path is wrong or not, and if the optimal path is wrong, continuing to advance; if an error occurs in the optimal path, for example, a collision occurs, that is, an undetected obstacle appears on the path, that is, the path error is sensed by a collision sensor arranged on the support 1, the position of the obstacle is obtained, a signal is fed back to the singlechip, path adjustment is realized through an error adjustment mechanism, and the action of the speed reduction motor 4 is controlled by the singlechip.
The error adjustment mechanism is as follows: since the route is determined based on the returned distance value from the robot to the surrounding object, a nearest obstacle is added again in the collision direction after an error occurs, and then the route is re-determined, so that the jam at the position is prevented.
6) And storing the obtained optimal path in real time and updating the optimal path in real time, so that the optimal path information is ensured to be used later after the exploration is finished.
7) Detecting and judging the current position, if the current position reaches the maximum wifi signal distance, placing a signal station by the mechanical arm 6, storing the current position, and detecting the current position again; otherwise, continuing to advance;
8) Returning according to the final path after reaching the end point, storing the final path, and recycling the signal station;
9) According to a series of data transmitted to an environment sensor with a function according to needs, such as a harmful gas sensor, a temperature and humidity sensor and the like, which are arranged on the robot support 1, the driving path of the robot is estimated, and corresponding estimation results are stored so as to be convenient for primary understanding of the working condition of the detected environment;
wherein, the evaluation process is: integrating the data acquired by various environmental sensors into a risk degree value of each sensor according to different preset weight ratios of the corresponding sensors, wherein the risk degree value is an evaluation result. Generally, the risk level value for each sensor is equal to the sum of the product of the data valid values for all environmental sensors and the weight ratio occupied by that environmental sensor; the evaluation result also comprises environment information of the explored position, temperature and humidity, road surface flatness and the like.
In each step, if a fire disaster occurs, after the flame sensor on the support 1 generates a signal, the singlechip models the ambient temperature according to data transmitted by a plurality of thermistors around the support 1, and the modeling process is as follows:
according to the installation position of the thermistor and the transmitted data, surrounding space points can be abstracted into a one-dimensional matrix containing x, y, z coordinates and temperature T, an isothermal line is established by finding out points with the same temperature, a normal direction unit vector of the isothermal line is calculated according to the coordinates of the points, and a calculation formula of a temperature gradient is obtained by the normal direction unit vector:
wherein n is a normal direction unit vector, and DeltaT is a temperature difference in the normal direction; Δn is the distance between the two isotherms corresponding to the temperature difference Δt in the normal direction thereof.
And automatically advancing to the lowest temperature direction, namely the gradient minimum direction according to a calculation formula of the temperature gradient.
In the above step 4), as shown in fig. 4, the algorithm fusion processing procedure is as follows:
4.1 Classifying the received data;
4.2 The multi-information fusion algorithm of the multi-core function support vector machine is adopted, and a linear kernel function, a polynomial kernel function and a radial basis kernel function method are mainly adopted to analyze signals such as sound, force (according to signals transmitted by a collision sensor), temperature, light and the like in the running process of the robot by the support vector machine, so as to obtain a global optimal solution, obtain an optimal kernel function, and send the optimal kernel function to a fusion rule center to obtain a fusion rule set;
wherein the fusion rule set may be expressed as Γ= { r 1 ,…,r n -where r i (i=1, 2, …, n) represents the decision rule of the i-th sensor.
4.3 Obtaining an expected risk value according to the fusion rule set, and if the expected risk value is the minimum expected risk value R (Γ), entering the next step; otherwise, returning to the step 4.1);
wherein the minimum expected risk value R (Γ) is:
wherein H is 1 Indicating the presence of a faulty target; h 0 Indicating that no fault target exists; l (L) ij As H j True time is judged as H i Probability of (2); p (P) j To assume H j Is (i, j=0, 1).
4.4 And (3) obtaining n groups of judgment rules through the minimum expected risk function solution, namely outputting a feature set, and completing multi-information fusion processing to select an optimal path.
In the above step 9), the evaluation effect of the present invention was verified by experiments: according to the weight ratio of harmful gas 1 concentration to harmful gas 2 concentration to road surface flatness to temperature to humidity, the factors are scaled by using an Analytic Hierarchy Process (AHP), and a judgment matrix and a weight ratio are obtained. The following is a weighting method, scale is shown in table 1.
TABLE 1
| Dangerous value (A) | Road surface flatness | Temperature (temperature) | Humidity of the water | Concentration of harmful gas 2 | Concentration of harmful gas 1 |
| Road surface flatness | 1 | 5 | 7 | 1 | 1/3 |
| Temperature (temperature) | 1/5 | 1 | 2 | 1/5 | 1/8 |
| Humidity of the water | 1/7 | 1/2 | 1 | 1/7 | 1/9 |
| Concentration of harmful gas 2 | 1 | 5 | 7 | 1 | 1/3 |
| Concentration of harmful gas 1 | 3 | 8 | 9 | 3 | 1 |
With judgment matrixThe maximum characteristic root, [ B, D ] of MATLAB software is obtained]=eig (a) (all eigenvalues of matrix a are found, diagonal matrix D is constructed, and eigenvectors of a are found, to construct column vectors of B), then there are:
B＝0.3697+0.0000i -0.0645+0.2358i -0.0645-0.2358i -0.2806+0.0000i -0.7071+0.0000i 0.0906+0.0000i -0.0633-0.0182i -0.0633+0.0182i 0.2303+0.0000i -0.0000+0.0000i 0.0595+0.0000i -0.0063-0.0620i -0.0063+0.0620i-0.1231+0.0000i 0.0000+0.0000i 0.3697+0.0000i -0.0645+0.2358i -0.0645-0.2358i-0.2806+0.0000i 0.7071+0.0000i 0.8455+0.0000i 0.9339+0.0000i 0.9339+0.0000i0.8799+0.0000i -0.0000+0.0000i
D＝5.1141+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i -0.0177+0.7618i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i -0.0177-0.7618i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i -0.0786+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i 0.0000+0.0000i -0.0000+0.0000i
the maximum feature root λ= 5.1141,
the consistency index ci= (λ -n)/(n-1) = (5.1141-5)/(5-1) =0.0285,
average random uniformity index RI (5) =1.12,
the random consistency ratio cr=ci/ri=0.0255 <0.10, so the weight is the normalized result of the feature vector corresponding to the maximum feature root after passing the consistency test:
W＝[0.2131 0.0522 0.0343 0.2131 0.4873]
namely, the weight ratio of the concentration of the harmful gas 1 is 0.4873, the weight ratio of the concentration of the harmful gas 2 is 0.2131, the weight ratio of the road surface evenness is 0.2131, the weight ratio of the temperature is 0.0522, and the weight ratio of the humidity is 0.0343.
The hazard value a=harmful gas 1 concentration value 0.4873 +harmful gas 2 concentration value 0.2131 +road surface flatness 0.2131 +temperature value 0.0522 +humidity value 0.0343.
In summary, the path processed by the fusion algorithm is converted into the motor control information to drive the motor of the speed reducer to act, and the default driving state is forward after the robot starts to run.
The foregoing embodiments are only illustrative of the present invention, and the structure, dimensions, placement and shape of the components may vary, and all modifications and equivalents of the individual components based on the teachings of the present invention should not be excluded from the scope of protection of the present invention.
Claims (6)
1. The intelligent exploration robot detection method based on multi-sensor fusion detection is characterized by comprising a support, wherein a crawler chassis and a damping mechanism are respectively arranged on two sides of the support, and the damping mechanism is positioned on the crawler chassis; the two sides of the support are also provided with thermistors, flame sensors, ultrasonic sensors and infrared sensors, and the thermistors are evenly distributed on the two sides of the support; the infrared sensor and the ultrasonic sensor are arranged at the front non-shielding position of the support; the flame sensor is arranged at a position on the support, which can be in contact with air with larger and more open area; the single-chip microcomputer is used for receiving signals transmitted by the collision sensor, the thermistor, the flame sensor, the ultrasonic sensor and the infrared sensor and controlling the action of the speed-reducing motor; the signal station performs information interaction with the singlechip; a mechanical arm is arranged at the top of the support, and the mechanical arm is controlled by the singlechip to work; a power supply is also arranged in the support;
each damping mechanism comprises a damping spring and a damping arm; the two damping arms are arranged in a crossing way, the crossing points are movably connected to the side wall of the support, the two end parts of the damping arms above are arranged in the track in a sliding way, and the track is fixed on the side wall of the support; the two ends of the damping arm positioned below are respectively connected with the thrust wheel; an arc-shaped arm which is integrally formed with each damping arm is arranged above the intersection point, the two arc-shaped arms are correspondingly arranged to form a semicircular arc, and the end parts of the two arc-shaped arms are provided with the damping springs;
the support is provided with a plurality of environment sensors, and the plurality of environment sensors are evenly distributed close to the top of the support; the laser ranging sensors are arranged at the edges of the two sides of the support and used for monitoring the position of the robot, and data acquired by the environment sensor and the laser ranging sensors are transmitted to the singlechip;
the detection method comprises the following steps:
1) Setting starting and ending positions;
2) The singlechip controls the gear motor to work and keeps the advancing state of the robot;
3) Each infrared sensor, each laser ranging sensor, each ultrasonic sensor and each environmental sensor perform data acquisition according to a preset rule, and the acquired data are transmitted into the singlechip;
4) Carrying out algorithm fusion processing according to the received data, obtaining an optimal path, feeding back to the singlechip, and returning to the step 2) to control the action of the gear motor so as to drive the driving wheel to advance in the optimal path;
5) Judging whether the optimal path is wrong or not, and if the optimal path is wrong, continuing to advance; if the optimal path is wrong, path adjustment is realized through an error adjustment mechanism, and then the action of the speed reduction motor is controlled by the singlechip;
6) Storing the obtained optimal path in real time and updating the optimal path in real time, so that the optimal path information is ensured to be used later after the exploration is finished;
7) Detecting and judging the current position, if the current position reaches the maximum wifi signal distance, placing a signal station, storing the current position, and detecting the current position again; otherwise, continuing to advance;
8) Returning according to the final path after reaching the end point, storing the final path, and recycling the signal station;
9) According to the data transmitted by the environment sensor arranged on the robot support, estimating the running path of the robot, and storing a corresponding estimation result so as to be convenient for primary understanding of the working condition of the detected environment;
in the step 4), the algorithm fusion processing process is as follows:
4.1 Classifying the received data;
4.2 The multi-information fusion algorithm of the multi-core function support vector machine is adopted, and a linear kernel function, a polynomial kernel function and a radial basis function method are mainly adopted to carry out support vector machine analysis on sound, force, temperature and light signals in the running process of the robot, obtain a global optimal solution, obtain an optimal kernel function and send the optimal kernel function to a fusion rule center to obtain a fusion rule set;
wherein the fusion rule set is expressed as Γ= { r 1 ,…,r n -where r i (i=1, 2, …, n) represents the decision rule of the i-th sensor;
4.3 Obtaining an expected risk value according to the fusion rule set, and if the expected risk value is the minimum expected risk value R (Γ), entering the next step; otherwise, returning to the step 4.1);
wherein the minimum expected risk value R (Γ) is:
wherein H is 1 Indicating the presence of a faulty target; h 0 Indicating that no fault target exists; l (L) ij As H j True time is judged as H i Probability of (2); p (P) j To assume H j I, j=0, 1;
4.4 Obtaining n groups of judgment rules through the minimum expected risk function solution, namely outputting a feature set, and completing multi-information fusion processing to select an optimal path;
the error adjustment mechanism is as follows: since the route is determined based on the returned distance value from the robot to the surrounding object, a nearest obstacle is added again in the collision direction after an error occurs, and then the route is re-determined, so that the jam at the position is prevented.
2. The intelligent exploration robot detection method based on multi-sensor fusion detection of claim 1, wherein the intelligent exploration robot detection method is characterized by comprising the following steps: if a fire disaster occurs, after a flame sensor on the support generates signals, the singlechip models the ambient temperature according to data transmitted by a plurality of thermistors around the support:
according to the installation position of the thermistor and the transmitted data, surrounding space points are abstracted into a one-dimensional matrix containing x, y, z coordinates and temperature T, an isothermal line is established by finding out points with the same temperature, a normal direction unit vector of the isothermal line is calculated according to the coordinates of the points, and a calculation formula of a temperature gradient is obtained by the normal direction unit vector:
wherein n is a normal direction unit vector, and DeltaT is a temperature difference in the normal direction; Δn is the distance between the two isotherms corresponding to the temperature difference Δt in the normal direction thereof;
and automatically advancing to the lowest temperature direction, namely the gradient minimum direction according to a calculation formula of the temperature gradient.
3. The intelligent exploration robot detection method based on multi-sensor fusion detection of claim 1, wherein the intelligent exploration robot detection method is characterized by comprising the following steps: a plurality of signal stations are arranged in the support, and each signal station is provided with a magnet device; the front end of the mechanical arm is provided with a micro electromagnet sucker controlled by the singlechip, and the micro electromagnet sucker is used for sucking the magnet device fixed on the signal station so as to place or take away the signal station.
4. The intelligent exploration robot detection method based on multi-sensor fusion detection of claim 1, wherein the intelligent exploration robot detection method is characterized by comprising the following steps: each crawler chassis comprises a guide wheel, a crawler, a riding wheel, a driving wheel, a transmission shaft, a thrust wheel, a driving shaft, a toothed belt and a synchronous pulley; an output shaft of the speed reducing motor is coaxially connected with the synchronous pulley, and the synchronous pulley is connected with the transmission shaft through the toothed belt; the transmission shaft is connected with the driving shaft through a synchronous pulley transmission mechanism, and the driving wheel is arranged on the driving shaft; the guide wheels, the riding wheels, the driving wheels and the supporting wheels are all arranged on the support and form a frame of the crawler chassis, and the crawler is positioned at the outer side of the frame and driven to move by the driving wheels; the guide wheel is positioned in front of the support, the driving wheel is positioned behind the support, and the height between the center position of the guide wheel and the ground is larger than that between the center position of the driving wheel and the ground; the supporting wheel is positioned between the guide wheel and the driving wheel, and the height between the central position of the supporting wheel and the ground is larger than that between the central position of the guide wheel and the ground; the thrust wheel is arranged on the damping mechanism.
5. The intelligent exploration robot detection method based on multi-sensor fusion detection of claim 4, wherein: the synchronous pulley transmission mechanism comprises the synchronous pulley and a belt; the transmission shaft is coaxially provided with the synchronous pulleys, the driving shaft is also provided with the synchronous pulleys, and the two synchronous pulleys are in transmission connection through the belt.
6. The intelligent exploration robot detection method based on multi-sensor fusion detection of claim 1, wherein the intelligent exploration robot detection method is characterized by comprising the following steps: the support is made of carbon fiber plates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711089041.2A CN107891423B (en) | 2017-11-08 | 2017-11-08 | Intelligent exploration robot based on multi-sensor fusion detection and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711089041.2A CN107891423B (en) | 2017-11-08 | 2017-11-08 | Intelligent exploration robot based on multi-sensor fusion detection and detection method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107891423A CN107891423A (en) | 2018-04-10 |
| CN107891423B true CN107891423B (en) | 2024-01-26 |
Family
ID=61804385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711089041.2A Active CN107891423B (en) | 2017-11-08 | 2017-11-08 | Intelligent exploration robot based on multi-sensor fusion detection and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107891423B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108633400B (en) * | 2018-04-24 | 2021-05-18 | 黑龙江八一农垦大学 | Automatic corn planter |
| CN109158807A (en) * | 2018-08-31 | 2019-01-08 | 南京涵铭置智能科技有限公司 | A kind of Intelligent welding mechanical arm |
| CN109394476B (en) * | 2018-12-06 | 2021-01-19 | 上海神添实业有限公司 | Method and system for automatic intention recognition of brain muscle information and intelligent control of upper limbs |
| CN110187410B (en) * | 2019-06-18 | 2021-05-18 | 武汉中海庭数据技术有限公司 | Human body detection device and method in automatic driving |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2001078951A1 (en) * | 2000-04-13 | 2001-10-25 | Zhimin Lin | Semi-optimal path finding in a wholly unknown environment |
| CN1945351A (en) * | 2006-10-21 | 2007-04-11 | 中国科学院合肥物质科学研究院 | Robot navigation positioning system and navigation positioning method |
| KR100770655B1 (en) * | 2006-12-19 | 2007-10-29 | 동일파텍주식회사 | A fire fighting robot |
| CN102161202A (en) * | 2010-12-31 | 2011-08-24 | 中国科学院深圳先进技术研究院 | Full-view monitoring robot system and monitoring robot |
| CN103770634A (en) * | 2014-01-22 | 2014-05-07 | 中国人民解放军国防科学技术大学 | Single-person all-terrain crawler-type moving carrier |
| CN106003064A (en) * | 2016-06-17 | 2016-10-12 | 上海工程技术大学 | Complex-environment multi-sensor intelligent detection robot |
| CN106882282A (en) * | 2015-12-15 | 2017-06-23 | 西北农林科技大学 | A kind of crawler-type traveling chassis of hydraulic-driven |
| CN107272705A (en) * | 2017-07-31 | 2017-10-20 | 中南大学 | A kind of multiple neural network controlling planning method of robot path under intelligent environment |
| CN207606847U (en) * | 2017-11-08 | 2018-07-13 | 石家庄铁道大学 | A kind of Intelligent exploration robot based on Multi-sensor Fusion detection |
-
2017
- 2017-11-08 CN CN201711089041.2A patent/CN107891423B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2001078951A1 (en) * | 2000-04-13 | 2001-10-25 | Zhimin Lin | Semi-optimal path finding in a wholly unknown environment |
| CN1945351A (en) * | 2006-10-21 | 2007-04-11 | 中国科学院合肥物质科学研究院 | Robot navigation positioning system and navigation positioning method |
| KR100770655B1 (en) * | 2006-12-19 | 2007-10-29 | 동일파텍주식회사 | A fire fighting robot |
| CN102161202A (en) * | 2010-12-31 | 2011-08-24 | 中国科学院深圳先进技术研究院 | Full-view monitoring robot system and monitoring robot |
| CN103770634A (en) * | 2014-01-22 | 2014-05-07 | 中国人民解放军国防科学技术大学 | Single-person all-terrain crawler-type moving carrier |
| CN106882282A (en) * | 2015-12-15 | 2017-06-23 | 西北农林科技大学 | A kind of crawler-type traveling chassis of hydraulic-driven |
| CN106003064A (en) * | 2016-06-17 | 2016-10-12 | 上海工程技术大学 | Complex-environment multi-sensor intelligent detection robot |
| CN107272705A (en) * | 2017-07-31 | 2017-10-20 | 中南大学 | A kind of multiple neural network controlling planning method of robot path under intelligent environment |
| CN207606847U (en) * | 2017-11-08 | 2018-07-13 | 石家庄铁道大学 | A kind of Intelligent exploration robot based on Multi-sensor Fusion detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107891423A (en) | 2018-04-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107891423B (en) | Intelligent exploration robot based on multi-sensor fusion detection and detection method thereof | |
| CN106383519B (en) | Robot autonomous positioning charging system and method | |
| CA2973038A1 (en) | Combined inertial navigation and laser scanning coal shearer positioning device and method | |
| CN105119338A (en) | Mobile robot charging control system and method | |
| CN106338999A (en) | Intelligent following anti-collision dolly and anti-collision method thereof | |
| CN207606847U (en) | A kind of Intelligent exploration robot based on Multi-sensor Fusion detection | |
| CN202216696U (en) | Coal mine disaster relief robot navigation device based on information integration | |
| CN103092202B (en) | Robot track positioning method and robot track positioning system | |
| CN205081492U (en) | Mobile robot control system that charges | |
| CN107014375B (en) | Indoor positioning system and method with ultra-low deployment | |
| CN107073711A (en) | A kind of robot follower method | |
| CN108415417A (en) | A kind of robot obstacle-avoiding system and method based on the prediction of barrier motion state | |
| CN105044754A (en) | Mobile platform outdoor positioning method based on multi-sensor fusion | |
| CN207928093U (en) | A kind of office chair for capableing of automatic homing | |
| CN105277593A (en) | Mobile robot based indoor smell source positioning method | |
| CN209634595U (en) | A kind of novel mine detection robot with automatic obstacle avoiding function | |
| CN107764229A (en) | Intelligent amendment type surface evenness detection device | |
| CN205426176U (en) | Coal -winning machine positioner that inertial navigation and laser scanning fuse | |
| CN206378299U (en) | Preceding anticollision test system based on distance test | |
| CN103257353A (en) | Wireless locating method of mixed-type multiple-mobile-robot system | |
| CN107664312A (en) | Oil smoke induction system, method and device | |
| CN100573372C (en) | Make up the method and apparatus of all-directional mobile device movement model | |
| CN206847566U (en) | Novel and multifunctional tunnel deformation robot measurement | |
| Distante et al. | An application of mobile robotics for olfactory monitoring of hazardous industrial sites | |
| CN103996264B (en) | A kind of intelligent movable security system based on WLAN communication |
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 |