The Robot’s Camera Setup Became an Input

A robot policy trained with fixed cameras can become brittle when a camera moves or disappears. VersaCamVLA separates representation of the camera set from action learning, converting variable views into a fixed-size latent. Target-view prediction and sampling from natural wrist-camera motion supply training signals and pose diversity. At deployment, a lightweight spatial encoder adds the scene tokens to a base vision-language-action policy without explicit 3D sensing or novel-view rendering. On RoboTwin, LIBERO and a real-robot platform, the authors report better results than prior methods and direct multi-view baselines across changed camera counts and unseen poses. The abstract supplies no single aggregate gain; the contribution is tested flexibility in the visual interface.