New neural network optimizes lunar lander trajectories

Researchers have developed a novel Optimality-Informed Neural Network (OINN) approach for optimizing the trajectory of a lunar lander during its powered descent. This method hard-codes necessary conditions of optimality, such as Pontryagin's minimum principle and the Hamilton-Jacobi-Bellman equation, directly into the network architecture. The OINN approach was tested against an independently solved boundary-value problem and Monte Carlo simulations, demonstrating close agreement and consistently small residuals, indicating its potential for real-time deployment with fixed computational costs. AI

IMPACT This research could lead to more efficient and reliable autonomous landing systems for spacecraft.

RANK_REASON The cluster contains a research paper detailing a new methodology for trajectory optimization. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• Lunar Lander
• Moon landing
• Optimality-Informed Neural Network
• Pontryagin's maximum principle

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →