Policy gradient approach for dynamic sensor tasking applied to space situational awareness

Physics & Astronomy

• situational awareness 100%
• learning 42%
• reinforcement 41%
• catalogs 39%
• gradients 38%
• sensors 36%
• optical measuring instruments 22%
• orbits 14%
• estimates 11%
• performance 8%
• simulation 7%

Engineering & Materials Science

• Reinforcement learning 59%
• Uncertainty 37%
• Optical sensors 35%
• Object tracking 31%
• Markov processes 31%
• Sensors 30%
• Orbits 26%

Earth & Environmental Sciences

• sensor 40%
• policy 35%
• reinforcement 29%
• learning 24%
• decision process 18%
• simulation 6%
• test 5%