RoboGolf: Mastering Real-World Minigolf with a Reflective Multi-Modality Vision-Language Model
This work addresses the challenge of embodied intelligence in minigolf for robotics, but it appears incremental as it builds on existing VLM methods with specific enhancements.
The authors tackled the problem of mastering real-world minigolf, which requires spatial, kinodynamic, and reflective reasoning, by introducing RoboGolf, a VLM-based framework with dual-camera perception and closed-loop action refinement, achieving results demonstrated through offline inference on recorded trajectories.
Minigolf is an exemplary real-world game for examining embodied intelligence, requiring challenging spatial and kinodynamic understanding to putt the ball. Additionally, reflective reasoning is required if the feasibility of a challenge is not ensured. We introduce RoboGolf, a VLM-based framework that combines dual-camera perception with closed-loop action refinement, augmented by a reflective equilibrium loop. The core of both loops is powered by finetuned VLMs. We analyze the capabilities of the framework in an offline inference setting, relying on an extensive set of recorded trajectories. Exemplary demonstrations of the analyzed problem domain are available at https://jity16.github.io/RoboGolf/