Autonomous Visual Rendering using Physical Motion
This work addresses the challenge of robotic visual rendering for tasks like drawing, offering a more generalizable approach that reduces dependence on task-specific robots, though it appears incremental in improving control methods.
This paper tackles the problem of enabling robots to recreate visual information through physical motion, such as drawing, by using ergodicity as a control objective to translate planar visual input directly into motion without preprocessing. It achieves comparable results to existing methods while reducing algorithmic complexity and demonstrates applicability across multiple robotic systems, including physical drawings with the Baxter robot.
This paper addresses the problem of enabling a robot to represent and recreate visual information through physical motion, focusing on drawing using pens, brushes, or other tools. This work uses ergodicity as a control objective that translates planar visual input to physical motion without preprocessing (e.g., image processing, motion primitives). % or human-generated training data (i.e., machine learning). We achieve comparable results to existing drawing methods, while reducing the algorithmic complexity of the software. We demonstrate that optimal ergodic control algorithms with different time-horizon characteristics (infinitesimal, finite, and receding horizon) can generate qualitatively and stylistically different motions that render a wide range of visual information (e.g., letters, portraits, landscapes). In addition, we show that ergodic control enables the same software design to apply to multiple robotic systems by incorporating their particular dynamics, thereby reducing the dependence on task-specific robots. Finally, we demonstrate physical drawings with the Baxter robot.