Generative grasp synthesis from demonstration using parametric mixtures
This work addresses grasp synthesis for robotics, offering incremental improvements in speed and success rates.
The paper tackles the problem of learning generative models for grasp synthesis from demonstration by proposing a parametric formulation that is computationally faster and achieves at least 10% higher grasp success rates in simulated experiments.
We present a parametric formulation for learning generative models for grasp synthesis from a demonstration. We cast new light on this family of approaches, proposing a parametric formulation for grasp synthesis that is computationally faster compared to related work and indicates better grasp success rate performance in simulated experiments, showing a gain of at least 10% success rate (p < 0.05) in all the tested conditions. The proposed implementation is also able to incorporate arbitrary constraints for grasp ranking that may include task-specific constraints. Results are reported followed by a brief discussion on the merits of the proposed methods noted so far.