Stochastic Optimization for Adapting Behaviors of Exploration Agents

Proposed missions to explore comets and moons will encounter environments that are hostile and unpredictable. A successful explorer must be able to adapt to a wide range of possible operating solutions to survive. Constructing special-purpose behaviors requires information about the environment, which is not available a priori for these missions. Instead, we propose an explorer that uses a flexible problemsolver with a significant capacity to adapt its behavior. More specifically, the explorer uses stochastic optimization techniques to continually adapt its behavior while limiting the cost of behavior exploration. With Adaptive Problem Solving, we use search techniques to enable a spacecraft to continually adapt its environment-specific behavior in-situ.