CN102646169A - Method for calculating mean free path (MFP) of exploration rover against complex terrain environment - Google Patents

Abstract

The invention discloses a method for calculating a mean free path (MFP) of an exploration rover against complex terrain environment, which comprises selecting a locating point and a running direction of the exploration rover; judging passibility of a projection region of the exploration rover body; and calculating the MFP. The invention really reflects the exploration environment of the exploration rover by establishing a digital elevation map (DEM) of star terrain, and can be applied to complex terrain environment having various obstacles (such as slopes and irregular obstacles), and the calculation accuracy is not affected by the complex degree of terrain. The invention calculates the MFP by rotating the DEM of the star terrain at a fixed angle interval and straightly moving along the fixed direction on each map, simplifies the algorithm complicacy, and simultaneously improves the simulation rate. The method adopts the slope and elevation difference as a termination judgment condition of a single MFP calculation, and comprehensively considers climbing capacity and obstacle clearance capacity of the exploration rover.

Description

A kind of inspection prober mean free path computing method to complex-terrain

Technical field

The invention belongs to aerospace field, relate to the computing method of inspection prober mean free path, specially refer to a kind of inspection prober mean free path computing method that are directed to complex-terrain.

Background technology

Celestial body input inspection prober is the key link in survey of deep space field outside the moon, Mars etc.Mean free path (Mean Free Path; MFP) be one and run into the probability of barrier and the relevant statistical measures of ability that breaks the barriers with inspection prober; It is defined as: in the specific landform, and before must changing travel direction, the mean path of inspection prober institute ability straight line moving.Its main application has following two aspects: the travelling performance of inspection prober is estimated in (1) under specific landform, for inspection prober size Selection, suspension design, chassis design provide foundation.(2) under the situation that inspection prober is confirmed, estimate the quality of landform passability, for the selection of touchdown area and the evaluation of touchdown area safety coefficient provide foundation.At present; The computing formula of mean free path is proposed in " Rover Chassis Evaluation and Design Optimizations Using the RCET " by people such as S.Michaud; At first, its size is described with diameter to the barrier that hinders the inspection prober roaming in certain zone; Add up the number of different-diameter scope barrier in this zone then, realize the calculating of mean free path based on following formula:

$x=\frac{1-\frac{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HDA0000155735990000072.png"\; state="\{\{state\}\}">b</figure-callout>}}{2}\underset{i=i_0}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{D}_i{N}_i-\frac{1}{2}\underset{i=i_0}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{D}_i^2{N}_i}{b\underset{i=i_0}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{N}_i+\underset{i=i_0}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{D}_i{N}_i}---\left(1\right)$

In the formula: b is a width of the carbody, and D is for hindering the barrier diameter of inspection prober roaming, and N is that every square metre of scope interior diameter is the number of the barrier of D～D+ δ D; The barrier that subscript i representative is different.

In practical application; Because celestial body surface environment more complicated; There is following defective in existing mean free path computing formula: the situation of only having considered only to exist in the subdued topography similar round barrier; The circular barrier that all is reduced to certain diameter as if the barrier with obstruction inspection prober roamings such as the rock that comes in every shape, hole moats is handled, and brings certain error inevitably; When having landform such as slope, the mean free path computing formula has only been considered the obstacle climbing ability of inspection prober, and has ignored the checking to its climbing capacity, and the method for promptly calculating mean free path through formula can not be considered other travelling performances outside the obstacle climbing ability.

Summary of the invention

For solving the problems referred to above that prior art exists; The present invention proposes a kind of inspection prober mean free path computing method to the complex-terrain environment; Its computational accuracy does not receive the influence of degree with a varied topography, can be applicable to exist in the terrain environment of all kinds barrier; In mean free path computation process, not only to consider the obstacle climbing ability of inspection prober, also will consider the climbing capacity of inspection prober, and can add other travelling performance indexs arbitrarily; Reduce algorithm complex, effectively improve computing velocity.

Technical scheme of the present invention is: a kind of inspection prober mean free path computing method that are directed to the complex-terrain environment may further comprise the steps:

The selection of A, inspection prober anchor point and travel direction

A1, generation celestial body landform digital elevation figure Digital Elevation Map, DEM

The terrain data of certain celestial body surf zone of being described by digital elevation model is one group of tri-vector finite sequence, uses the formal description of function to be: V _i=(X _i, Y _i, Z _i) (i=1,2,3 ..., n), wherein, X _i, Y _iBe planimetric coordinates, Z _iBe X _i, Y _iCorresponding elevation; The scope that realizes planimetric coordinates and elevation of in Matlab software, at first programming is extracted, and promptly seeks directions X minimum value X _MinWith maximal value X _Max, Y direction minimum value Y _MinWith maximal value Y _Max, Z direction minimum value Z _MinWith maximal value Z _MaxThen, confirm the bulk of each pixel representative in the map and the height value of different gray-scale value representatives, set up the digital elevation figure that describes with gray level image, wherein, the gray-scale value of 0-255 and Z _Min-Z _MaxBetween be linear mapping relations;

A2, selection travel direction

Mean free path need expend very long simulation time for the mean value of the free path repeatedly selecting reference position and travel direction at random and obtain; In order to reduce algorithm complex; Improve simulation velocity; Selection is rotated celestial body landform digital elevation figure DEM at interval by fixed angle, and the method for on every width of cloth figure, all going along fixed-direction then averages free path and calculates, car body view field complexity of calculation when avoiding that choice direction is gone at random;

A3, random selecting point go

After definite fixedly travel direction; In every width of cloth celestial body landform digital elevation figure DEM, select the central point of starting point at random, in celestial body landform digital elevation figure DEM, extract the landform altitude data in the car body drop shadow spread around the central point according to the car body size as inspection prober starting;

B, inspection prober car body view field passability are judged

The passability judgement comprises two aspects, the one, and gradient restriction, another is the vertical height restriction, both are all the single free path and calculate the decision condition that stops; If inspection prober is wide is b, and obstacle climbing ability is that vertical height is h, and the gradient that can ascend is restricted to θ; Wherein, step B1 is that the gradient is judged, step B2 is that vertical height is judged; Can pass through when both are all inspection prober through result of determination, judge that then car body view field can pass through;

Landform altitude data in B1, the selection car body drop shadow spread; Utilize least square method to carry out the space plane match; The grade information of the space plane parameter acquiring car body view field that obtains according to match; According to the driveability of inspection prober, the gradient is greater than θ can not pass through this zone, calculates one of decision condition that stops as the single free path;

B2, in the zone of the anterior overall width b*h/tan θ of view field, search for minimum height value and maximum elevation value, if both differences then think to exceed the obstacle climbing ability scope greater than h, promptly can not pass through, calculate another decision condition of termination as the single free path;

C, mean free path are calculated

During each simulation calculation free path, running into craspedodrome to be terminal point through locating, and obtains the longest air line distance that single can go, and repeatedly statistics back sum-average arithmetic just can be obtained mean free path.

Compared with prior art, beneficial effect of the present invention and benefit are:

1, the mean free path computing method of the present invention's proposition; Through setting up celestial body landform digital elevation figure DEM; The acquisition environment of true reflection inspection prober; Can be applicable to exist in the complex-terrain environment of all kinds obstacle (for example slope and irregular slalom), and computational accuracy does not receive the influence of degree with a varied topography.

2, the present invention is through rotating celestial body landform digital elevation figure DEM figure at interval by fixed angle, and the method for on every width of cloth figure, all keeping straight on along fixed-direction averages free path calculating, reduced algorithm complex, improved simulation velocity simultaneously.

3, the present invention utilizes the landform altitude data of inspection prober car body view field; Carry out plane fitting and obtain view field's grade information; Car body view field search for minimum height value and maximum elevation value and do poor; The gradient and elevation value difference are calculated the decision condition that stops as the single free path, taken all factors into consideration the climbing capacity and the obstacle climbing ability of inspection prober.

4, the present invention can arbitrarily increase new single free path calculating termination decision condition in step B, for example, considers that the degree of roughness of landform surpasses a certain threshold value etc.So the present invention can consider other travelling performance indexs outside the obstacle climbing ability in mean free path is calculated.

Description of drawings

The present invention has accompanying drawing 15 width of cloth, wherein:

Fig. 1 is the celestial body landform digital elevation figure DEM in certain lunar surface landform zone.

Fig. 2 is 5 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Fig. 3 is 10 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Fig. 4 is 20 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Fig. 5 is 30 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Fig. 6 is 45 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Fig. 7 is 90 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Fig. 8 is 180 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Fig. 9 is 315 ° of degree celestial body landform digital elevation figure DEM rotation examples.

Figure 10 is the car body view field that extracts along working direction.

Figure 11 is that grade information is obtained in the space plane match.

Figure 12 is that minimum height value and maximum elevation value hunting zone are confirmed in the car body view field.

Figure 13 is the result that 315 ° of rotation celestial body landform digital elevation figure DEM proceed to 3000 free path simulation calculation.

Figure 14 is whenever at a distance from 5 ° of free path statistics curves that carry out 10000 walkings.

Figure 15 is a process flow diagram of the present invention.

Embodiment

Below in conjunction with technical scheme and accompanying drawing the present invention is further specified.The size of the inspection prober of present embodiment is long to be 1.5m, wide b=1.0m, and gradient θ is restricted to 20 °, and obstacle climbing ability (vertical direction) h is 20cm.

The selection of A, inspection prober device anchor point and travel direction

The first step is with the 60*60m in moonscape Apollo 15 touch-down zones ²Terrain data be example, selecting the space length of every pixel representative is 60mm, can set up the celestial body landform digital elevation figure DEM of 1000*1000 resolution shown in Figure 1, the minimum height value Z in this zone _MinFor-1.7m, maximum elevation value Z _MaxBe 1.45m, the gray level of 0-255 among the corresponding celestial body landform digital elevation figure DEM, promptly every gray level is represented the floor level of 12.35mm.

In second step, the rotation of celestial body landform digital elevation figure DEM can arbitrarily be adjusted in mean free path computation process at interval, and is more little at interval, and mean free path is calculated accurate more, and this instance finally is chosen as 5 ° of intervals, carried out 72 rotations.Fig. 2-9 is depicted as 5 °, 10 °, 20 °, 30 °, 45 °, 90 °, 180 °, the 315 ° celestial body landform digital elevation figure DEM rotation examples of picking out.

The 3rd step; Fixed vertical upwards is travel direction; In every width of cloth celestial body landform digital elevation figure DEM, select the central point of starting point at random, in celestial body landform digital elevation figure DEM, extract the central point lunar surface altitude figures in the car body drop shadow spread on every side according to car body size length * wide=1.5m * 1.0m as inspection prober starting.Shown in figure 10.

B, inspection prober car body view field passability are judged

The first step; Utilize least square method that the lunar surface altitude figures in the car body drop shadow spread is carried out the space plane match; Shown in figure 11; The space line equation that obtains is z=ax+by+c; According to the grade information of this view field of space plane parameter acquiring, the climbing capacity of the gradient

inspection prober is 20 ° then, if the gradient θ that calculates shows promptly that greater than 20 ° car body view field can not pass through.

In second step, in the zone of overall width * 20/tan shown in Figure 12 (20 °), search for minimum height value and maximum elevation value, if the difference of the two judges promptly that greater than the 20cm that the inspection prober obstacle climbing ability limits car body view field can not pass through.

If the difference of the value of slope that calculates and maximum, minimum height value all within the scope that inspection prober can be realized, judges that then car body view field can pass through.

C, mean free path are calculated

In simulation process; To serve as that 72 width of cloth celestial body landform digital elevation figure DEM that rotate at interval carry out 10000 walkings respectively with 5 °; Running into craspedodrome to be terminal point through locating, and obtains the longest air line distance that at every turn can go, and 72*10000 result calculated sued for peace; And then average, can obtain the mean free path of final inspection prober in this landform zone.For example, shown in Figure 13 in the celestial body landform digital elevation figure of 315 ° of rotations DEM, to proceed to 3000 times the free path result calculated, the black line track that to be car body pass by along direction vertically upward among the figure.Figure 14 is the result of calculation curve of mean free path on all directions, add with average after can to obtain the mean free path of selected inspection prober in this landform zone be 9.23m.

Claims (1)

1. inspection prober mean free path computing method that are directed to the complex-terrain environment is characterized in that: may further comprise the steps:

The selection of A, inspection prober anchor point and travel direction

A1, generation celestial body landform digital elevation figure Digital Elevation Map, DEM

The terrain data of certain celestial body surf zone of being described by digital elevation model is one group of tri-vector finite sequence, uses the formal description of function to be: V _i=(X _i, Y _i, Z _i) (i=1,2,3 ..., n), wherein, X _i, Y _iBe planimetric coordinates, Z _iBe X _i, Y _iCorresponding elevation; The scope that realizes planimetric coordinates and elevation of in Matlab software, at first programming is extracted, and promptly seeks directions X minimum value X _MinWith maximal value X _Max, Y direction minimum value Y _MinWith maximal value Y _Max, Z direction minimum value Z _MinWith maximal value Z _MaxThen, confirm the bulk of each pixel representative in the map and the height value of different gray-scale value representatives, set up the digital elevation figure that describes with gray level image, wherein, the gray-scale value of 0-255 and Z _Min-Z _MaxBetween be linear mapping relations;

A2, selection travel direction

Mean free path need expend very long simulation time for the mean value of the free path repeatedly selecting reference position and travel direction at random and obtain; In order to reduce algorithm complex; Improve simulation velocity; Selection is rotated celestial body landform digital elevation figure DEM at interval by fixed angle, and the method for on every width of cloth figure, all going along fixed-direction then averages free path and calculates, car body view field complexity of calculation when avoiding that choice direction is gone at random;

A3, random selecting point go

After definite fixedly travel direction; In every width of cloth celestial body landform digital elevation figure DEM, select the central point of starting point at random, in celestial body landform digital elevation figure DEM, extract the landform altitude data in the car body drop shadow spread around the central point according to the car body size as inspection prober starting;

B, inspection prober car body view field passability are judged

The passability judgement comprises two aspects, the one, and gradient restriction, another is the vertical height restriction, both are all the single free path and calculate the decision condition that stops; If inspection prober is wide is b, and obstacle climbing ability is that vertical height is h, and the gradient that can ascend is restricted to θ; Wherein, step B1 is that the gradient is judged, step B2 is that vertical height is judged; Can pass through when both are all inspection prober through result of determination, judge that then car body view field can pass through;

Landform altitude data in B1, the selection car body drop shadow spread; Utilize least square method to carry out the space plane match; The grade information of the space plane parameter acquiring car body view field that obtains according to match; According to the driveability of inspection prober, the gradient is greater than θ can not pass through this zone, calculates one of decision condition that stops as the single free path;

B2, in the zone of the anterior overall width b*h/tan θ of view field, search for minimum height value and maximum elevation value, if both differences then think to exceed the obstacle climbing ability scope greater than h, promptly can not pass through, calculate another decision condition of termination as the single free path;

C, mean free path are calculated

During each simulation calculation free path, running into craspedodrome to be terminal point through locating, and obtains the longest air line distance that single can go, and repeatedly statistics back sum-average arithmetic just can be obtained mean free path.

Images