Abstract:The probability that two satellites overlap in space at a specified instant of time is called their instantaneous collision probability. Assuming Gaussian uncertainties and spherical satellites, this probability is the integral of a Gaussian distribution over a sphere. This paper shows how to compute the probability using an established numerical procedure called characteristic function inversion. The collision probability in the short-term encounter scenario is also evaluated with this approach, where the instant at which the probability is computed is the time of closest approach between the objects. Python and R code is provided to evaluate the probability in practice. Overall, the approach has been established for over fifty years, is implemented in existing software, does not rely on analytical approximations, and can be used to evaluate two and three dimensional collision probabilities.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to calculate the instantaneous collision probability (ICP) of satellites at a specific time point. Specifically, when the distance between two satellites in space is less than a certain threshold, they may collide. Assuming that the position uncertainty of the satellites is a Gaussian distribution and the satellites are spherical, this probability can be calculated by integrating the Gaussian distribution on a sphere. ### Main problems 1. **Calculation of instantaneous collision probability**: How to accurately calculate the instantaneous collision probability between two satellites at a given time point. 2. **Collision probability in short - term encounter scenarios**: Calculate the collision probability of two satellites at the closest moment in the short - term encounter scenario (STES). ### Solutions The paper proposes to use the Characteristic Function Inversion (CFI) method to calculate the instantaneous collision probability. CFI is a numerical method that can effectively handle the generalized chi - square distribution, which is derived from the probability distribution of the squared separation distance. ### Specific steps 1. **Define the model and probability expression**: - Let \(R(t)\) be the relative position of two satellites at time \(t\), and assume that \(R(t)\sim N(\mu(t),\Sigma(t))\), where \(\mu(t)\) is the mean vector of the relative position and \(\Sigma(t)\) is the covariance matrix. - The condition for a collision to occur is \(R'R < q^{2}\), where \(q\) is the collision distance threshold. - The instantaneous collision probability \(p_{c}(t)=P(R'R < q^{2})\) can be expressed as a triple integral: \[ p_{c}(t)=\int_{\{r|r'r < q^{2}\}}\exp\left(-\frac{1}{2}(r - \mu)'\Sigma^{-1}(r-\mu)\right)dr \] 2. **Generalized chi - square distribution**: - The squared separation distance \(R'R\) follows a generalized chi - square distribution, that is, \(R'R=\sum_{i = 1}^{3}\lambda_{i}Y_{i}^{2}\), where \(Y_{i}\sim\chi^{2}_{1}(\delta_{i})\) is a non - central chi - square distribution, \(\lambda_{i}\) is an eigenvalue of the covariance matrix \(\Sigma\), and \(\delta_{i}\) is a non - central parameter. 3. **Characteristic function inversion**: - Use the characteristic function inversion formula proposed by Davies (1973, 1980) to calculate the instantaneous collision probability: \[ p_{c}(t)=\frac{1}{2}-\int_{-\infty}^{\infty}\text{Im}\left(\phi(s)\exp(-i s q^{2})/(2\pi s)\right)ds \] - where \(\phi(s)=\prod_{i = 1}^{3}\phi_{i}(\lambda_{i}s)\), and \(\phi_{i}(s)\) is the characteristic function of the non - central chi - square distribution. 4. **Software implementation**: - Provide Python and R code to implement the above calculation method and verify the results through Monte Carlo simulation. ### Paper contributions - Propose a new method for calculating the instantaneous collision probability using the characteristic function inversion method. - Without the need for analytical approximation, it only depends on numerical integration and is applicable to Gaussian or Gaussian mixture models. - Provide an efficient numerical implementation and compare it with existing methods.

Computing the Instantaneous Collision Probability between Satellites using Characteristic Function Inversion

Probability of Collision of satellites and space debris for short-term encounters: Rederivation and fast-to-compute upper and lower bounds

Satellite conjunction analysis and the false confidence theorem

Distance and Collision Probability Estimation from Gaussian Surface Models

Collision probability estimation for small unmanned aircraft systems

Short-Term Collision Probability Caused by Debris Cloud

Collision Criterion for Two Satellites on Keplerian Orbits

Generating Probability Distributions on Intervals and Spheres with Application to Finite Element Method.

Reduction of Space Debris Collision Prediction Uncertainty Based on Q-Sat Precise Orbit

An Analytic Method of Space Debris Cloud Evolution and Its Collision Evaluation for Constellation Satellites

Collision Probability Prediction and Orbit Maneuvering Probability Determination of Non-Cooperative Space Object Orbit

Hybrid Method of Uncertainty Propagation for Near-Earth Conjunction Analysis

Satellite Safe Margin: Fast solutions for Conjunction Analysis

Adjusting the catastrophic collision model using outer probability measure theory

A Study on the Collision Probability of Satellite-Based ADS-B Messages Based on Homogeneous Poisson Process

Generating Probability Distributions on Intervals and Spheres: Convex Decomposition

A Rapid Method for Calculating Maximal and Minimal Inter-Satellite Distances

Density-based in-orbit collision risk model valid for any impact geometry

Modeling Medium-Term Debris Cloud of Satellite Breakup via Probabilistic Method

Fast Collision Probability Estimation for Automated Driving using Multi-circular Shape Approximations

Jacobian Spheroids, Shallow Encounters, and the Keyhole Map