A method and system for intelligently identifying high-risk terrain using thermal inertia

By calculating the thermal inertia of the star table and establishing a Gaussian hybrid model to identify the star table traps, the problem of non-geometric risk terrain identification in extraterrestrial detection is solved, and the accurate prediction of the terrain passability is achieved to ensure the safe movement of the detector.

CN115290074BActive Publication Date: 2024-11-05BEIJING INST OF CONTROL ENG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210761819.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-06-29
Publication Date
2024-11-05
Estimated Expiration
2042-06-29

AI Technical Summary

Technical Problem

The prior art is difficult to effectively identify and predict the looseness and hardness of the non-geometric risk terrain in extraterrestrial detection, resulting in wheel slipping, sinking or permanently trapping the detector.

Method used

By calculating the thermal inertia of the star table and establishing a Gaussian hybrid model, the expected maximization algorithm is used to estimate the model parameters and identify the star table traps, thereby predicting the passability of the terrain.

Benefits of technology

It realizes accurate prediction of the terrain passability, ensures the safe movement and detection accuracy of the detector, and has self-learning ability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115290074B_ABST
    Figure CN115290074B_ABST
Patent Text Reader

Abstract

The present invention discloses a high-risk terrain intelligent identification method and system using thermal inertia, wherein the method comprises: obtaining the star surface thermal inertia according to the star surface temperature; using the star surface thermal inertia to establish a Gaussian mixture model for identifying star surface traps. The present invention can predict the passability of the terrain and ensure accurate and safe detection of the detector.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of extraterrestrial exploration technology, and in particular relates to a method and system for intelligently identifying high-risk terrain using thermal inertia. Background Art

[0002] Extraterrestrial exploration has gradually extended from the nearest moon to increasingly distant celestial bodies such as Mars and asteroids, and the exploration methods have gradually progressed from flybys and orbits to landings, patrols and sample returns.

[0003] During extraterrestrial patrols, there are often many risks caused by dangerous terrain environments on the surface of the planet. These risks can be divided into geometric risks and non-geometric risks. Geometric risks refer to obstacles such as large rocks, craters, and steep slopes, while non-geometric risks refer to soft soil and thin weathering layers. In order to improve mobility and safely drive to the target area, the rover should be able to autonomously perceive the surface environment. For geometric risks on the surface of the planet, they can be identified through vision and lidar, allowing the rover to avoid these obstacles and improve mobility. However, for non-geometric risks on the surface of the planet, the above-mentioned traditional perception methods can only see the surface appearance and geometry, and it is difficult to predict the looseness and hardness of the terrain, which can easily cause the wheels to slip and sink, inhibiting mobility, and even permanently trapping the vehicle, such as the wheels of the "Spirit" rover eventually sank into the soft Martian soil and could not get out. Summary of the invention

[0004] The technical problem solved by the present invention is: to overcome the deficiencies of the prior art and provide a method and system for intelligently identifying high-risk terrain using thermal inertia, which can predict the passability of the terrain and ensure accurate and safe detection by the detector.

[0005] The purpose of the present invention is achieved through the following technical solutions: a method for intelligently identifying high-risk terrain using thermal inertia, comprising: obtaining star surface thermal inertia according to star surface temperature; and establishing a Gaussian mixture model using the star surface thermal inertia for identifying star surface traps.

[0006] In the above-mentioned high-risk terrain intelligent identification method using thermal inertia, the star table thermal inertia is obtained by the following formula:

[0007]

[0008] Among them, A is the albedo of the star, ε is the radiation coefficient, σ is the Boltzmann constant, and R sw is the solar shortwave radiation, R lw is the atmospheric long-wave radiation, T s is the surface temperature, I is the surface thermal inertia, t is the time, z′ is the ratio of depth to heat penetration depth, and T(z,t) is the surface temperature at time t and the ratio of depth to heat penetration depth z.

[0009] In the above-mentioned high-risk terrain intelligent identification method using thermal inertia, the ratio z′ of depth to heat penetration depth is: Where z is the depth and δ is the heat penetration depth.

[0010] In the above-mentioned high-risk terrain intelligent identification method using thermal inertia, the heat penetration depth δ is obtained by the following formula:

[0011]

[0012] Among them, k is thermal conductivity, ρ is density, c is specific heat capacity, and P is the day-night cycle period.

[0013] In the above-mentioned high-risk terrain intelligent identification method using thermal inertia, the Gaussian mixture model is obtained by the following formula:

[0014]

[0015] in, is the probability of the star catalog thermal inertia under the intermediate parameter θ, K is the number of models of the Gaussian mixture model, α n is the probability that the measured data belongs to the nth submodel, φ(I|θ n ) is the Gaussian distribution density function of the nth sub-model, θ is the intermediate parameter, θ n are the final estimates of the expectation and variance of the nth sub-model, I is the catalog thermal inertia, and n is the numerical label of the sub-model.

[0016] In the above-mentioned high-risk terrain intelligent identification method using thermal inertia, the intermediate parameter in, represents the expected final estimate of the nth sub-model, represents the final estimate of the covariance of the nth sub-model, Represents the final estimated value of the probability of the nth submodel occurring in the Gaussian mixture model.

[0017] A high-risk terrain intelligent identification system using thermal inertia includes: a first module for obtaining star surface thermal inertia according to star surface temperature; and a second module for establishing a Gaussian mixture model using the star surface thermal inertia for identifying star surface traps.

[0018] In the above-mentioned high-risk terrain intelligent identification system using thermal inertia, the star table thermal inertia is obtained by the following formula:

[0019]

[0020] Among them, A is the albedo of the star, ε is the radiation coefficient, σ is the Boltzmann constant, and R sw is the solar shortwave radiation, R lw is the atmospheric long-wave radiation, Ts is the surface temperature, I is the surface thermal inertia, t is the time, z′ is the ratio of depth to heat penetration depth, and T(z,t) is the surface temperature at time t and the ratio of depth to heat penetration depth z.

[0021] In the above-mentioned high-risk terrain intelligent identification system using thermal inertia, the ratio z′ of depth to heat penetration depth is: Where z is the depth and δ is the heat penetration depth.

[0022] In the above-mentioned high-risk terrain intelligent identification system using thermal inertia, the heat penetration depth δ is obtained by the following formula:

[0023]

[0024] Among them, k is thermal conductivity, ρ is density, c is specific heat capacity, and P is the day-night cycle period.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] (1) The present invention calculates the thermal inertia of the star surface based on the onboard and orbital instruments, establishes a Gaussian mixture model based on the estimated value of the thermal inertia, and uses the expectation maximization algorithm to estimate the parameters of the Gaussian mixture model, thereby obtaining a star surface trap recognition model, thereby predicting the passability of the terrain and ensuring accurate and safe detection by the detector.

[0027] (2) The present invention perceives and understands the star map and the physical properties under the star map by establishing a Gaussian mixture model, thereby predicting the passability of the terrain and having the ability of self-learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0029] Figure 1 It is a schematic diagram of the results of the Gaussian mixture model for trap identification provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0031] This embodiment provides a method for intelligently identifying high-risk terrain using thermal inertia, the method comprising the following steps:

[0032] The star surface thermal inertia is obtained according to the star surface temperature;

[0033] The Gaussian mixture model is established using the star-level thermal inertia to identify star-level traps.

[0034] Step 1: Calculation of thermal inertia of the star catalog

[0035] Thermal inertia determines the material's ability to resist temperature changes under periodic light exposure. Thermal inertia is defined as

[0036]

[0037] In the formula, k represents thermal conductivity, ρ represents density, c represents specific heat capacity, and the unit of thermal inertia is Jm -2 K -1 s -1 / 2 .

[0038] In order to calculate the thermal inertia of the star, the one-dimensional heat conduction equation is defined as follows:

[0039]

[0040] Where T(z,t) represents the surface temperature at time t and depth z, and z=0 represents the surface temperature.

[0041] The temperature of the surface of a star is determined by the balance between upwelling and downwelling from the sun, the atmosphere, and the ground. That is, it satisfies the following equation:

[0042]

[0043] Where A is the albedo of the star, ε is the radiation coefficient, σ is the Boltzmann constant, P is the day-night cycle, and R is sw represents solar shortwave radiation, R lw represents the atmospheric long-wave radiation, T s represents the surface temperature, Represents the surface temperature gradient value.

[0044] Let z' = z / δ, δ represents the heat penetration depth, satisfying Then formula (2) can be rewritten as

[0045]

[0046] Combining equation (3) and equation (4), we can get:

[0047]

[0048] Therefore, by measuring the star surface temperature using equation (4), the star surface thermal inertia I can be calculated, and a Gaussian mixture model can be established to identify star surface traps.

[0049] Step 2: Use the star-level thermal inertia I to establish a Gaussian mixture model for identifying star-level traps.

[0050] Equation (1) shows that thermal inertia is affected by the particle size, density and degree of cementation of the surface terrain material. Generally speaking, the surface terrain with higher degree of cementation, larger density and larger particle size has higher thermal inertia. For extraterrestrial patrol exploration missions, the larger the particles, the greater the soil strength. Similarly, denser and more compact sand bodies are easier to pass than fine sand and loose sand bodies. In addition, as the particles are bonded together, the terrain support force is stronger, making it easier to pass. Therefore, there is a strong positive correlation between traversability and thermal inertia. For surface traps, the increase in the depth of the ground sand layer leads to a decrease in both thermal inertia and traversability.

[0051] A mixture model is a probability model that can be used to represent a population distribution with K sub-distributions. In other words, the mixture model represents the probability distribution of the observed data in the population. It is a mixture distribution composed of K sub-distributions. Since the data corresponding to whether the patroller can pass through is independently distributed, that is, it satisfies the Gaussian distribution, the Gaussian mixture model is defined as follows:

[0052]

[0053] In the formula, represents the expected final estimate of the nth sub-model, represents the final estimate of the covariance of the nth sub-model, Represents the final estimated value of the probability of the nth submodel occurring in the Gaussian mixture model. is the probability of the star catalog thermal inertia under the intermediate parameter θ, K is the number of models of the Gaussian mixture model, α n is the probability that the measured data belongs to the nth submodel, φ(I|θ n ) is the Gaussian distribution density function of the nth sub-model, θ is the intermediate parameter, θ nis the final estimated value of the expectation and variance of the nth sub-model, I is the star catalog thermal inertia, and n is the numerical label of the sub-model. It should be understood that the Gaussian mixture model includes n sub-models.

[0054] The parameter θ in the Gaussian mixture model can be learned using the expectation maximization algorithm, as follows:

[0055] (1) Initialization parameters;

[0056] (2) Calculate each data I based on the current parameters j The probability γ from submodel n jn

[0057]

[0058] In the formula, φ(I j |θ n ) represents each thermal inertia data I j The probability of belonging to model n.

[0059] (3) Calculate the model parameter μ for the new iteration n , σ n and α n , where μ n , σ n and α n They represent the expected iterative estimate of each sub-model, the difference in the covariance iterative estimate of each sub-model, and the probability iterative estimate occurring in the mixed model.

[0060]

[0061] Repeat steps (2) and (3) until convergence, and finally obtain the parameters of the Gaussian mixture model Therefore, based on the established Gaussian mixture model, the thermal inertia observation value of the target area can be input to obtain the probability of whether the current area is a trap, thereby judging the passability.

[0062] This embodiment also provides a high-risk terrain intelligent identification system using thermal inertia, the system comprising: a first module for obtaining the star surface thermal inertia according to the star surface temperature; a second module for establishing a Gaussian mixture model using the star surface thermal inertia for identifying star surface traps.

[0063] The trap identification model established in this embodiment has the following results: Figure 1 As shown, the simulation results show that the recognition accuracy is 90.72%.

[0064] The present invention calculates the thermal inertia of the star surface based on the onboard and orbital instruments, establishes a Gaussian mixture model based on the estimated value of the thermal inertia, and uses the expectation maximization algorithm to estimate the parameters of the Gaussian mixture model, thereby obtaining a star surface trap recognition model, thereby predicting the passability of the terrain and ensuring accurate and safe detection by the detector; the present invention perceives and understands the star surface and the physical properties under the star surface by analyzing the thermal inertia characteristics of the star surface, thereby predicting the passability of the terrain, and has the ability of self-learning.

[0065] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A high-risk terrain intelligent identification method using thermal inertia, characterized in that include: The star surface thermal inertia is obtained according to the star surface temperature; The Gaussian mixture model is established using the star-level thermal inertia to identify star-level traps. The star table thermal inertia is obtained by the following formula: Among them, A is the albedo of the star, ε is the radiation coefficient, σ is the Boltzmann constant, and R sw is the solar shortwave radiation, R lw is the atmospheric longwave radiation, Τ s is the surface temperature, I is the surface thermal inertia, t is the time, z' is the ratio of depth to heat penetration depth, T(z,t) is the surface temperature at time t and the ratio of depth to heat penetration depth z; The ratio of depth to heat penetration depth z' is: Where z is the depth and δ is the heat penetration depth; The heat penetration depth δ is obtained by the following formula: Among them, k is thermal conductivity, ρ is density, c is specific heat capacity, and P is the diurnal cycle period; The Gaussian mixture model is obtained by the following formula: in, is the probability of the star catalog thermal inertia under the intermediate parameter θ, K is the number of models of the Gaussian mixture model, α n is the probability that the measured data belongs to the nth submodel, φ(I|θ n ) is the Gaussian distribution density function of the nth sub-model, θ is the intermediate parameter, θ n is the final estimate of the expectation and variance of the nth sub-model, I is the catalog thermal inertia, and n is the numerical label of the sub-model; Intermediate parameters in, represents the expected final estimate of the nth sub-model, represents the final estimate of the covariance of the nth sub-model, Represents the final estimated value of the probability of the nth submodel occurring in the Gaussian mixture model.

2. A high-risk terrain intelligent identification system using thermal inertia, characterized in that include: The first module is used to obtain the star surface thermal inertia according to the star surface temperature; The second module is used to establish a Gaussian mixture model using the thermal inertia of the star table to identify star table traps; The star table thermal inertia is obtained by the following formula: Among them, A is the albedo of the star, ε is the radiation coefficient, σ is the Boltzmann constant, and R sw is the solar shortwave radiation, R lw is the atmospheric longwave radiation, Τ s is the surface temperature, I is the surface thermal inertia, t is the time, z' is the ratio of depth to heat penetration depth, T(z,t) is the surface temperature at time t and the ratio of depth to heat penetration depth z; The ratio of depth to heat penetration depth z' is: Where z is the depth and δ is the heat penetration depth; The heat penetration depth δ is obtained by the following formula: Among them, k is thermal conductivity, ρ is density, c is specific heat capacity, and P is the diurnal cycle period; The Gaussian mixture model is obtained by the following formula: in, is the probability of the star catalog thermal inertia under the intermediate parameter θ, K is the number of models of the Gaussian mixture model, α n is the probability that the measured data belongs to the nth submodel, φ(I|θ n ) is the Gaussian distribution density function of the nth sub-model, θ is the intermediate parameter, θ n is the final estimate of the expectation and variance of the nth sub-model, I is the catalog thermal inertia, and n is the numerical label of the sub-model; Intermediate parameters in, represents the expected final estimate of the nth sub-model, represents the final estimate of the covariance of the nth sub-model, Represents the final estimated value of the probability of the nth submodel occurring in the Gaussian mixture model.