Posts tagged Vishnu Desaraju
Experience-driven Predictive Control with Robust Constraint Satisfaction under Time-Varying State Uncertainty

Vehicle dynamics and model parameters are rarely well known in practice. Robot controllers are further challenged by actuator constraints and noisy state estimates. This paper presents a controller that efficiently learns vehicle dynamics while at the same time efficiently ensuring that the vehicle satisfies state and input constraints in the presence of state uncertainty. Tracking error improvements are shown while maintaining constraint satisfaction and flying a quadrotor through a windy environment.

Read More
Experience-Based Models of Surface Proximal Aerial Robot Flight Performance in Wind

We developed a strategy for an aerial robot to build a map of the wind strength and direction in its operating environment by aggregating its flight experiences into a non-parametric model. By considering this wind map in path planning, the aerial robot is able to move between points with greater safety and energy efficiency.

Read More