XGBoost and transformer models trained without temperature data outperform feed-forward neural networks in predicting BioTac sensor outputs when using force and contact inputs.
Using Tactile Sensing to Improve the Sample Efficiency and Performance of Deep Deterministic Policy Gradients for Simulated in-Hand Manipulation Tasks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Optimizing BioTac Simulation for Realistic Tactile Perception
XGBoost and transformer models trained without temperature data outperform feed-forward neural networks in predicting BioTac sensor outputs when using force and contact inputs.