Learning Trajectories for Visual-Inertial System Calibration via Model-based Heuristic Deep Reinforcement Learning
This addresses the need for precise calibration in visual-inertial systems, which typically require complex manual motions, by automating trajectory generation for robotics applications, representing an incremental improvement over existing methods.
The paper tackled the problem of generating favorable trajectories for visual-inertial system calibration by modeling it as a Markov decision process and using model-based deep reinforcement learning with particle swarm optimization, resulting in lower calibration errors compared to random or handcrafted trajectories while maintaining similar or shorter path lengths.
Visual-inertial systems rely on precise calibrations of both camera intrinsics and inter-sensor extrinsics, which typically require manually performing complex motions in front of a calibration target. In this work we present a novel approach to obtain favorable trajectories for visual-inertial system calibration, using model-based deep reinforcement learning. Our key contribution is to model the calibration process as a Markov decision process and then use model-based deep reinforcement learning with particle swarm optimization to establish a sequence of calibration trajectories to be performed by a robot arm. Our experiments show that while maintaining similar or shorter path lengths, the trajectories generated by our learned policy result in lower calibration errors compared to random or handcrafted trajectories.