A group of researchers from Caltech and the Jet Propulsion Laboratory(JPL) have developed an emergency system for autonomous space vehicles that optimizes diagnostic and response processes in real time, helping to avoid accidents on future space missions.
Safety in autonomous space vehicles at Caltech
This project was named s-FEAST(Safe Fault Estimation via Active Sensing Tree Search), and allows spacecraft to identify and correct problems without human intervention, which is essential in critical situations where there is no time to communicate with Earth.
[post_relacionado]Professor Soon-Jo Chung’s lab at Caltech has developed s-FEAST technology, an algorithm that simulates multiple possible approaches/futures in real time, helping spacecraft identify failures and correct them without stopping the mission. It works like a “muscle scanner,” carefully testing spacecraft systems and taking corrective action before major problems arise.
In space, autonomous vehicles can face unexpected challenges, such as propellant failures or collisions with objects, which can jeopardize multibillion-dollar missions. So far, engineers have used redundant systems and safe modes to deal with these problems. However, these solutions add cost and weight to the spacecraft.
Implementation and execution of tests
The system has been tested in a multi-spacecraft “dynamics simulator,” where a simulated spacecraft floats over a nearly frictionless surface to mimic space conditions. According to James Ragan, a Ph.D. student involved in the project, the algorithm quickly analyzes with numerous possible futures, allowing the spacecraft to evaluate“the best trajectory course of action” to avoid danger, such as collisions with asteroids.
Autonomous spacecraft testing for collision avoidance. Source: Caltech
In addition, its use in spacecraft space vehicless-FEAST has been adapted to test ground vehicles, with equally promising results. This technology is expected to make space missions and autonomous exploration both on Earth and in space safer and more cost-effective.
This work, published in Science Roboticscould transform the aerospace industry’s approach to autonomous vehicle design.
Follow us on social networks and don’t miss any of our publications!
YouTube LinkedIn Facebook Instagram X
Source and photo: Caltech