Event
Paper on unified planning framework for tactile sorting released
1 reports · 1 sources
Zhuoyi Lu and other authors submitted a paper proposing Haptic Sorter, a unified model-based geometric framework that combines robotic tactile perception, modeling, and manipulation planning for manipulation when object shape and pose are unknown. The framework uses Bayesian optimization to guide tactile exploration, approximates geometric boundaries with superellipses, describes quasi-static robot-object interaction with an adaptive manipulation potential that encodes object geometry, and proposes an online ordinary differential equation implementation based on model prediction and tactile feedback for real-time pose inference. The authors state that the system was deployed on a 2D robotic sorting task and that, by varying object geometry, they verified the framework's robustness and generalization in simulation and on a real multi-arm setup.
Report timeline
- Haptic Sorter: A Unified Planning Framework for Online Shape Estimation and Real-Time Pose Inference
The authors propose Haptic Sorter, a unified model-based geometric framework that integrates robotic haptic perception, modeling, and manipulation planning for manipulation when object shape and pose are unknown. The framework uses Bayesian optimization to guide haptic exploration and approximates geometric boundaries with superellipses, describes quasi-static robot-object interaction with an adaptive manipulation potential that encodes object geometry, and proposes an online ordinary differential equation based on model prediction and tactile feedback for real-time pose inference. The system is deployed on a 2D robotic sorting task, varying object geometries to validate the framework's robustness and generalizability in both simulation and a real-world multi-arm setup.
arXiv cs.RO ·