# Learning active tactile perception through belief-space control

arXiv: https://arxiv.org/abs/2312.00215 · HTML: https://mlsanity.com/paper/2312.00215/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 23 references that are in the map.
- First posted on arXiv 30 Nov 2023.

## Builds on (most cited first, 10 of 10)

- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [On the Properties of Neural Machine Translation: Encoder-Decoder Approaches](https://mlsanity.com/paper/1409.1259/) (2014)
- [Learning Latent Dynamics for Planning from Pixels](https://mlsanity.com/paper/1811.04551/) (2018)
- [robosuite: A Modular Simulation Framework and Benchmark for Robot Learning](https://mlsanity.com/paper/2009.12293/) (2020)
- [Backprop KF: Learning Discriminative Deterministic State Estimators](https://mlsanity.com/paper/1605.07148/) (2016)
- [Learning to Perform Physics Experiments via Deep Reinforcement Learning](https://mlsanity.com/paper/1611.01843/) (2016)
- [Multimodal Sensor Fusion with Differentiable Filters](https://mlsanity.com/paper/2010.13021/) (2020)
- [SwingBot: Learning Physical Features from In-hand Tactile Exploration for Dynamic Swing-up Manipulation](https://mlsanity.com/paper/2101.11812/) (2021)
- [On 'A Kalman Filter-Based Algorithm for IMU-Camera Calibration: Observability Analysis and Performance Evaluation'](https://mlsanity.com/paper/1311.4769/) (2013)
- [On Training Recurrent Networks with Truncated Backpropagation Through Time in Speech Recognition](https://mlsanity.com/paper/1807.03396/) (2018)

## Built on by (most cited first, 1 of 1)

- [AdaptiGraph: Material-Adaptive Graph-Based Neural Dynamics for Robotic Manipulation](https://mlsanity.com/paper/2407.07889/) (2024)

## Abstract (arXiv)

Robots operating in an open world will encounter novel objects with unknown physical properties, such as mass, friction, or size. These robots will need to sense these properties through interaction prior to performing downstream tasks with the objects. We propose a method that autonomously learns tactile exploration policies by developing a generative world model that is leveraged to 1) estimate the object's physical parameters using a differentiable Bayesian filtering algorithm and 2) develop an exploration policy using an information-gathering model predictive controller. We evaluate our method on three simulated tasks where the goal is to estimate a desired object property (mass, height or toppling height) through physical interaction. We find that our method is able to discover policies that efficiently gather information about the desired property in an intuitive manner. Finally, we validate our method on a real robot system for the height estimation task, where our method is able to successfully learn and execute an information-gathering policy from scratch.