CN110488837B - A method for confirming pseudo-source of gas source - Google Patents

Abstract

A method for confirming a pseudo-source of a gas source, which relates to the technical field of gas source sourcing, and in order to solve the problem of low positioning accuracy of the gas source in the prior art, the method includes step 1: obtaining the position of the gas source and using it as the origin, and paralleling the origin with the origin. Establish a confirmation area coordinate system on the coordinate axis of the search area; Step 2: Establish a cost map of the surrounding environment for the obstacle, and perform expansion processing on the obstacle; Step 3: According to the confirmation area coordinate system and the cost map in the surrounding environment of the obstacle Plan the confirmation area boundary, select the rectangular area as the gas source confirmation area, make the robot move, and construct the confirmation area boundary; Step 4: The robot moves independently along the confirmation area boundary three times, and the gas concentration at the confirmation area boundary is measured by the gas sensor; The formula determines the authenticity of the gas source, and the present invention is based on a statistical method, including two parts: the gas source confirmation rule and the confirmation area boundary division rule, and the gas source confirmation accuracy is high.

Description

A method for confirming pseudo-source of gas source

technical field

The invention relates to the technical field of gas source seeking, in particular to a method for confirming a false source of a gas source.

Background technique

In the obstacle scene, the accumulation of gas on the windward side of the obstacle, the large measurement error of the sensor, and the strong discontinuity of the gas propagation will all form a false source, which will affect the accuracy of the gas source positioning. Therefore, when searching for a gas source using various methods, it is necessary to use the gas source pseudo-source confirmation algorithm to determine whether the obtained gas source location actually contains a gas source.

At present, there are few methods for the robot to confirm the gas source through its own wind speed/wind direction sensor and gas sensor, mainly including the recently measured position sequence method, the confirmation method based on machine learning, and the mass flux divergence method. The main problems of these methods are: the calculation is cumbersome and complicated, the requirements for the gas concentration and the environmental flow field are relatively strict, and they are greatly affected by the turbulent environment.

SUMMARY OF THE INVENTION

The purpose of the present invention is to propose a method for confirming a pseudo-source of a gas source, aiming at the problem of low positioning accuracy of the gas source in the prior art.

The technical scheme adopted by the present invention in order to solve the above-mentioned technical problems is: a gas source pseudo-source confirmation method, comprising the following steps:

Step 1: Obtain the position of the gas source and use it as the origin, and establish the confirmation area coordinate system with the origin parallel to the coordinate axis of the search area;

Step 2: Build a cost map of the surrounding environment for the obstacle, and perform expansion processing on the obstacle;

Step 3: Plan and confirm the area boundary according to the coordinate system of the confirmation area and the cost map in the surrounding environment of the obstacle, select the rectangular area as the gas source confirmation area, make the robot move, and construct the confirmation area boundary;

Step 4: The robot independently moves three circles along the boundary of the confirmation area, and the gas concentration at the boundary of the confirmation area is measured by the gas sensor;

Step 5: Determine the authenticity of the gas source according to the following formula,

O _S →( _NS ≥η)∧( _ns / _NS ≥ζ)

表示确认区域内不含气体源；式中符号→代表逻辑条件，∧表示逻辑与，∨表示逻辑或，N_S为一段时间内统计气体浓度超过设定的浓度阈值的次数，n_s为测得气体浓度超过阈值的事件发生时，粒子收敛位置处于确认区域范围内的次数，η和ζ为经验阈值。In the formula, O _S indicates that there is a gas source in the confirmed area,

Indicates that there is no gas source in the confirmation area; in the formula, the symbol → represents the logical condition, ∧ represents the logical AND, ∨ represents the logical OR, N _S is the number of times that the statistical gas concentration exceeds the set concentration threshold in a period of time, and n _s is the measured value. When the gas concentration exceeds the threshold, the number of times that the particle convergence position is within the range of the confirmed region, η and ζ are empirical thresholds.

Further, the expansion of obstacles in the second step is implemented based on the local cost map of the ROS navigation module.

Further, when the obstacle is expanded in the second step, the expansion cost formula in the expansion area around the obstacle is:

_Cost = 253×exp(-1.0×k×(dr))

In the formula, k is the expansion scale factor, d is the distance from the obstacle, and r is the radius of the inscribed circle of the robot.

Further, the gas source confirmation area in the step 3 is represented as:

L _conf =max{6max{σ _x ,σ _y },2l _d }

表示确认区域坐标系下x方向的最小值，

表示确认区域坐标系下x方向的最大值，

表示确认区域坐标系下y方向的最小值,

表示确认区域坐标系下y方向的最大值。In the formula, S _conf is the confirmation area, L _conf is the length or width of the rectangular confirmation area, σ _x and σ _y are the standard deviation of the particle distribution in the x and y directions when the particles converge, and l _d is the obstacle after the expansion process. The length or width of the unidirectional safety distance,

Indicates the minimum value in the x direction in the confirmation area coordinate system,

Indicates the maximum value in the x direction in the confirmation area coordinate system,

Indicates the minimum value in the y direction in the confirmation area coordinate system,

Indicates the maximum value in the y direction in the confirmation area coordinate system.

Further, the side length of the confirmation area of the confirmation area boundary first satisfies L _conf ≥ 6max{σ _x ,σ _y }, and then according to the requirement of no collision during robot navigation, the confirmation area boundary is expanded until all the boundaries meet the navigation requirements. .

The beneficial effects of the present invention are:

1. The present invention belongs to the category of single-robot gas source confirmation, and the cost is low;

2. The present invention is based on statistical methods, including two parts: the gas source confirmation rule and the confirmation area boundary division rule, the calculation speed is fast, and the gas source confirmation accuracy is high;

3. Since the existence of obstacles is considered when confirming the division of the area boundary, the invention is suitable for a complex environment with obstacles.

Description of drawings

FIG. 1a is a simulation environment of the present invention.

Figure 1b is a cost map of the present invention.

FIG. 2 is a schematic diagram of the evolution process of the confirmation area boundary planning according to the present invention.

FIG. 3 is a schematic diagram of the distribution of the gas field in the simulation experiment for confirming the pseudo-source of the gas source according to the present invention.

FIG. 4a is a schematic diagram 1 of a simulation experiment process for confirming a pseudo-source of a gas source according to the present invention.

FIG. 4b is a schematic diagram 2 of the simulation experiment process for the confirmation of the pseudo-source of the gas source according to the present invention.

FIG. 4c is a schematic diagram 3 of the simulation experiment process for confirming the pseudo-source of the gas source according to the present invention.

Detailed ways

Embodiment 1: This embodiment is described in detail with reference to FIG. 1. A method for confirming a gas source pseudo-source described in this embodiment includes the following steps:

Step 1: Obtain the position of the gas source and use it as the origin, and establish the confirmation area coordinate system with the origin parallel to the coordinate axis of the search area;

Step 2: Build a cost map of the surrounding environment for the obstacle, and perform expansion processing on the obstacle;

Step 3: Plan and confirm the area boundary according to the coordinate system of the confirmation area and the cost map in the surrounding environment of the obstacle, select the rectangular area as the gas source confirmation area, make the robot move, and construct the confirmation area boundary;

Step 4: The robot independently moves three circles along the boundary of the confirmation area, and the gas concentration at the boundary of the confirmation area is measured by the gas sensor;

Step 5: Determine the authenticity of the gas source according to the following formula,

O _S →( _NS ≥η)∧( _ns / _NS ≥ζ)

表示确认区域内不含气体源；式中符号→代表逻辑条件，∧表示逻辑与，∨表示逻辑或，N_S为一段时间内统计气体浓度超过设定的浓度阈值的次数，n_s为测得气体浓度超过阈值的事件发生时，粒子收敛位置处于确认区域范围内的次数，η和ζ为经验阈值。In the formula, O _S indicates that there is a gas source in the confirmed area,

Indicates that there is no gas source in the confirmation area; in the formula, the symbol → represents the logical condition, ∧ represents the logical AND, ∨ represents the logical OR, N _S is the number of times that the statistical gas concentration exceeds the set concentration threshold in a period of time, and n _s is the measured value. When the gas concentration exceeds the threshold, the number of times that the particle convergence position is within the range of the confirmed region, η and ζ are empirical thresholds.

The present invention uses the gas source identification rule based on the statistical method to confirm the gas source. The rule can be analogous to the process of identifying the gas source by humans. When humans confirm the existence of a gas source in a certain place, it is usually determined based on two common senses: firstly, at the place A certain gas can be clearly detected nearby, followed by the fact that the gas always floats from the direction of the location; when judging that there is no gas source in a certain place, it is generally based on the fact that a certain gas cannot be clearly detected near the location or the gas is not always Floating from the direction of the location.

The present invention is the last link of the gas source location search, and the present invention is based on the possible gas source locations that have been calculated through a certain gas source search algorithm. The technical scheme of the present invention is summarized as follows: take the calculated gas source position as the origin, establish a confirmation area coordinate system parallel to the coordinate axis of the search area; establish a cost map of the surrounding environment, and deal with the expansion of obstacles; according to the confirmation area coordinate system and surrounding The cost map planning in the environment confirms the boundary of the area; the robot independently moves three circles along the boundary of the confirmation area, and the gas concentration at the boundary of the area is confirmed by the gas sensor measurement; according to the gas source identification rules, the authenticity of the gas source is judged.

The technical solution specifically includes the following steps:

Step 1, take the calculated gas source position as the origin, and establish a confirmation area coordinate system parallel to the search area coordinate axis. The position of the gas source is calculated by the gas source search algorithm, which is the premise of implementing the present invention. Taking the search for the gas source based on the particle filter algorithm as an example, the position of the gas source is calculated by the particle convergence position. The search area refers to a sufficiently large area of the scene that is determined to include the real gas source.

Step 2: Build a cost map of the surrounding environment and expand the obstacles. In the obstacle scene, confirm that the area boundary may intersect with the obstacle, that is, the navigation target point is planned to be within the obstacle, so that the navigation cannot be completed, and the gas source is generally an obstacle. In order to prevent the robot from colliding with the gas source, in the real scene The presence of obstacles must be considered in the planning confirmation area. Therefore, a cost map of the surrounding environment must be established to inflate the obstacles.

The obstacle expansion processing is implemented based on the local cost map of the ROS navigation module. The Gazebo environment and cost map are shown in Figure 1. The costmap weighs the actual size of the robot to calculate the safe distance from obstacles. The area around the obstacle smaller than the radius of the inscribed circle of the robot is a high-cost area that is forbidden to enter, such as the light-colored center area in Figure 1b); the area around the obstacle larger than the radius of the inscribed circle of the robot and less than the expansion radius is the expansion area, such as In the dark area around the light-colored central area in Fig. 1b), the calculation formula of the expansion cost in this area is shown in formula (1).

_Cost = 253×exp(-1.0×k×(dr)) (1)

In the formula, k is the expansion scale factor, d is the distance from the obstacle, and r is the radius of the inscribed circle of the robot.

The coordinate system of the local cost map is the robot coordinate system, and the cost value of the point in the cost map can be obtained for a given target point in the cost map. Since the target point obtained by the robot when navigating in the scene is the coordinate of the origin of the robot coordinate system in the odometer coordinate system, that is, the point is located inside the robot, considering the actual shape of the robot, in order to avoid collision, in addition to the code corresponding to the test target point In addition to whether the value meets the no-collision requirement, the outline points of the robot should also be tested. In order to reduce the amount of calculation, the present invention simplifies the used mobile robot platform as a circle, and selects the distance between the front and rear, left and right directions of the target pose and the target point as the inscribed robot. The four points of the circle radius are used as test points. If any of the five points including the target point has a high cost value, the target point will be discarded and the next best target point will be selected. If the test points all meet the no-collision requirement, the target point will be sent to the navigation module for the robot to carry out Navigation controls. In the present invention, C _ost ≤ 80 is set to meet the requirement of no collision, otherwise the cost value is considered too high and collision may occur.

Step 3: Confirm the area boundary according to the coordinate system of the confirmed area and the cost map planning in the surrounding environment. In the present invention, considering that the mobile robot platform adopted is of relatively large quality, and it is difficult to achieve short-distance fine motion due to the influence of static friction, the rectangular motion trajectory of the robot is more in line with the requirements of the experimental conditions, and the difficulty of trajectory point selection can be reduced, and the rectangular area is selected as the The confirmation area of the gas source makes the robot move and constructs the boundary of the confirmation area. Taking the gas source search based on the particle filter algorithm as an example, the gas source position is calculated from the particle convergence position. The condition for the particle convergence is that the standard deviation of the particle in the x and y directions is less than the set value, and the confirmation area is expressed as:

L _conf =max{6max{σ _x ,σ _y },2l _d } (3)

In the formula, S _conf is the confirmation area, L _conf is the length or width of the rectangular confirmation area, σ _x and σ _y are the standard deviation of the particle distribution in the x and y directions when the particles converge, when other gas source search algorithms are used, σ _x and σ _y can be set according to the uncertainty of the gas source location of the algorithm. l _d is the unidirectional safety distance of the length or width of the obstacle after expansion treatment. In this way, the confirmation area can contain most of the particles, and the influence of obstacles is comprehensively considered, so as to avoid collision between the robot and obstacles during the gas source confirmation process.

The evolution process of confirming the area boundary planning is shown in Figure 2. Confirm that the side length of the area first satisfies L _conf ≥ 6max{σ _x ,σ _y } and then judge whether the corners of the confirmed area meet the no-collision requirement at this time. The points expand outward along the diagonal direction until all the corner points meet the no-collision requirements, and then re-expand into a rectangular confirmation area according to the size relationship of the four corner points. point, and judge again whether it meets the no-collision requirement. If not, the boundary of the confirmation area that will cause collision will expand outward as a whole, and the rest of the boundary will remain unchanged until the cost values of all boundary sampling points meet the navigation requirements.

Step 4, the robot independently moves three circles along the boundary of the confirmation area, and the gas concentration at the boundary of the confirmation area is measured by the gas sensor.

Step 5: Determine the authenticity of the gas source according to the gas source identification rule. The present invention uses the gas source identification rule based on the statistical method to confirm the gas source. The rule can be analogous to the process of identifying the gas source by humans. When humans confirm the existence of a gas source in a certain place, it is usually determined based on two common senses: firstly, at the place A certain gas can be clearly detected nearby, followed by the fact that the gas always floats from the direction of the location; when judging that there is no gas source in a certain place, it is generally based on the fact that a certain gas cannot be clearly detected near the location or the gas is not always Floating from the direction of the location. The above logic can be organized as follows:

O _S →( _NS ≥η)∧( _ns / _NS ≥ζ) (4)

为确认区域内不含气体源；式中符号→代表逻辑条件，∧表示逻辑与，∨表示逻辑或。N_S为一段时间内统计气体浓度超过设定的浓度阈值的次数，此浓度阈值为确认气体源专用的浓度阈值，n_s为通过各种方法搜索气体源，测得气体浓度超过阈值的事件发生时，计算得到气体源的位置处于确认区域范围内的次数。本发明以基于粒子滤波算法搜索气体源为例，则n_s为测得气体浓度超过阈值的事件发生时，粒子收敛位置处于确认区域范围内的次数。η和ζ为经验阈值。若满足(N_S≥η)∧(n_s/N_S≥ζ)，则认定为气体源，若满足(N_S＜η)∨(n_s/N_S＜ζ)则认定为气体伪源。In the formula, O _S indicates that there is a gas source in the confirmed area,

In order to confirm that there is no gas source in the area; in the formula, the symbol → represents the logical condition, ∧ represents the logical AND, and ∨ represents the logical OR. N _S is the number of times the gas concentration exceeds the set concentration threshold within a period of time, the concentration threshold is the concentration threshold dedicated to confirming the gas source, _ns is the gas source searched by various methods, and the measured gas concentration exceeds the threshold. , calculate the number of times that the position of the gas source is within the range of the confirmation area. The present invention takes the search for a gas source based on a particle filter algorithm as an example, and _ns is the number of times that the particle convergence position is within the range of the confirmation area when the measured gas concentration exceeds the threshold when the event occurs. η and ζ are empirical thresholds. If ( _NS ≥η)∧( _ns / _NS ≥ζ) is satisfied, it is identified as a gas source, and if ( _NS <η)∨( _ns / _NS <ζ) is met, it is identified as a gas pseudo source.

Embodiment: In order to further understand the content and characteristics of the present invention, the present invention is further described below with reference to the accompanying drawings and specific examples, taking the gas source search algorithm based on particle filtering as an example. Although the particle filter source search algorithm is used as an example, the implementation method of the present invention can also be extended to the pseudo-source confirmation process of other gas source search algorithms. The gas field distribution is calculated according to the simplified equation of atmospheric statistics, as shown in Figure 3.

The concrete steps of the present invention are as follows:

Step 1, take the calculated gas source position as the origin, and establish a confirmation area coordinate system parallel to the search area coordinate axis.

Step 2: Build a cost map of the surrounding environment and expand the obstacles. The calculation formula of the expansion cost in the expansion area around the obstacle is:

_Cost = 253×exp(-1.0×k×(dr)) (6)

In the formula, k is the expansion scale factor, d is the distance from the obstacle, and r is the radius of the inscribed circle of the robot. In the present invention, C _ost ≤ 80 is set to meet the requirement of no collision, otherwise the cost value is considered too high and collision may occur.

Step 3: Confirm the area boundary according to the coordinate system of the confirmed area and the cost map planning in the surrounding environment. In the present invention, the robot uses a rectangular motion trajectory, selects the rectangular area as the confirmation area of the gas source, makes the robot move, and constructs the boundary of the confirmation area. Taking the gas source search based on the particle filter algorithm as an example, the gas source position is calculated from the particle convergence position. The condition for particle convergence is that the standard deviation of the particles in the x and y directions is less than the set value. The gas source confirmation area can be expressed as:

L _conf =max{6max{σ _x ,σ _y },2l _d } (8)

In the formula, S _conf is the confirmation area, L _conf is the length or width of the rectangular confirmation area, σ _x and σ _y are the standard deviation of the particle distribution in the x and y directions when the particles converge, when other gas source search algorithms are used, σ _x and σ _y can be set according to the uncertainty of the gas source location of the algorithm. l _d is the unidirectional safety distance of the length or width of the obstacle after expansion treatment. The evolution process of confirming the regional boundary planning is shown in Figure 2.

Step 4, the robot independently moves three circles along the boundary of the confirmation area, and the gas concentration at the boundary of the confirmation area is measured by the gas sensor. In the simulation experiment, the position of the gas source is calculated by the source algorithm in advance. Taking the search for the gas source based on the particle filter algorithm as an example, the simulation process of the robot moving along the boundary of the confirmation area is shown in Figure 4. Figure 4a) is performed for each circle. The sampling results of the uniform distribution of particles before the gas source confirmation movement, the particles converge to the vicinity of the obstacle in Figure 4b), and Figure 4c) shows the particle state at the end of the gas source confirmation. It can be seen that the particles have not converged to the confirmation area.

Step 5: Determine the authenticity of the gas source according to the gas source identification rule. The present invention uses the gas source identification rule based on the statistical method to confirm the gas source, and the logic formula is as follows:

O _S →( _NS ≥η)∧( _ns / _NS ≥ζ) (9)

为确认区域内不含气体源；式中符号→代表逻辑条件，∧表示逻辑与，∨表示逻辑或。N_S为一段时间内统计气体浓度超过设定的浓度阈值的次数，此浓度阈值为确认气体源专用的浓度阈值，n_s为通过各种方法搜索气体源，测得气体浓度超过阈值的事件发生时，计算得到气体源的位置处于确认区域范围内的次数。本发明以基于粒子滤波算法搜索气体源为例，则n_s为测得气体浓度超过阈值的事件发生时，粒子收敛位置处于确认区域范围内的次数。η和ζ为经验阈值。In the formula, O _S indicates that there is a gas source in the confirmed area,

In order to confirm that there is no gas source in the area; in the formula, the symbol → represents the logical condition, ∧ represents the logical AND, and ∨ represents the logical OR. N _S is the number of times the gas concentration exceeds the set concentration threshold within a period of time, the concentration threshold is the concentration threshold dedicated to confirming the gas source, _ns is the gas source searched by various methods, and the measured gas concentration exceeds the threshold. , calculate the number of times that the position of the gas source is within the range of the confirmation area. The present invention takes the search for a gas source based on a particle filter algorithm as an example, and _ns is the number of times that the particle convergence position is within the range of the confirmation area when the measured gas concentration exceeds the threshold when the event occurs. η and ζ are empirical thresholds.

In the experiment, the gas source confirmation threshold was empirically set as η=0.4, ζ=0.6. According to the gas field distribution and the above steps, the final statistical data of the gas source confirmation process are: N _S =55, n _s =20, the total number of detections in 3 circles is 60, because 55/60>η, n _s /N _S =20 /55<ζ, it is determined by equation (10) that the particles converge to the pseudo-source. Combining with the gas field distribution in Figure 3, it can be seen that due to the accumulation of gas near the windward side of the obstacle, the pseudo-source is formed, resulting in the wrong convergence of the particles.

It should be noted that the specific embodiments are only explanations and descriptions of the technical solutions of the present invention, and cannot be used to limit the protection scope of the rights. Any changes made according to the claims and description of the present invention are only partial changes, which should still fall within the protection scope of the present invention.

Claims (4)

1. A gas source pseudo-source confirmation method comprising the steps of:

the method comprises the following steps: acquiring the position of a gas source, taking the position as an origin, and establishing a coordinate system of a confirmation area by taking the origin parallel to the coordinate axis of a search area;

step two: establishing a cost map of the surrounding environment of the obstacle, and performing expansion processing on the obstacle;

step three: planning a confirmed area boundary according to a confirmed area coordinate system and a cost map in the surrounding environment of the obstacle, selecting a rectangular area as a gas source confirmed area, enabling the robot to move, and constructing the confirmed area boundary;

step four: the robot independently moves three circles along the boundary of the confirmed area, and the concentration of the gas at the boundary of the confirmed area is measured through the gas sensor;

step five: the authenticity of the gas source is judged according to the following formula,

O_S→(N_S≥η)∧(n_s/N_S≥ζ)

in the formula, O_SIndicating that the gas source is contained within the validation region,

indicating that the confirmation area does not contain a gas source; in the formula, symbol → represents a logical condition, Λ represents a logical AND, V represents a logical OR, N_SCounting the times that the gas concentration exceeds a set concentration threshold value within a period of time, n_sThe times that the particle convergence position is within the range of the confirmation region when the event that the measured gas concentration exceeds the threshold occurs, wherein eta and zeta are empirical thresholds;

the method is characterized in that: the gas source identification area in the third step is expressed as:

L_conf＝max{6max{σ_x,σ_y},2l_d}

in the formula, S_confFor confirming the region, L_confFor rectangular confirmation of area length or width, σ_xAnd σ_yStandard deviations of the particle distributions in the x and y directions at convergence, l_dThe length or width of the barrier is the one-way safe distance after expansion treatment,

represents the minimum value in the x direction in the coordinate system of the confirmation area,

represents the maximum value in the x direction in the coordinate system of the confirmation area,

represents the minimum value in the y direction in the coordinate system of the confirmation area,

which represents the maximum value in the y-direction in the coordinate system of the validation area.

2. A gas source pseudo-source confirmation method as claimed in claim 1, wherein: and in the second step, the expansion processing of the obstacles is realized based on a local cost map of the ROS navigation module.

3. A gas source pseudo-source confirmation method as claimed in claim 1, wherein: when the barrier is subjected to expansion treatment in the second step, an expansion cost formula in the peripheral expansion area of the barrier is as follows:

C_ost＝253×exp(-1.0×k×(d-r))

in the formula, k is an expansion scale factor, d is the distance from the obstacle, and r is the radius of the inscribed circle of the robot.

4. A gas source pseudo-source confirmation method as claimed in claim 1, wherein: the side length of the confirmation region boundary satisfies L_conf≥6max{σ_x,σ_yAnd then expanding the boundaries of the confirmed area according to the requirement of no collision during the robot navigation until all the boundaries meet the navigation requirement.

Images