Learning State-Dependent Losses for Inverse Dynamics Learning

Morse, Kristen; Das, Neha; Lin, Yixin; Wang, Austin S.; Rai, Akshara; Meier, Franziska

doi:10.1109/iros45743.2020.9341701

Cited by 6 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mahler et al [15] demonstrated that LSTM networks can outperform the GP-based method for modeling inverse dynamics. While many works focus on the model architecture, Morse et al [16] apply meta-learning to obtain state-dependent loss functions, which demonstrates another viable approach.…”

Section: Related Workmentioning

confidence: 99%

Flexible-Joint Manipulator Trajectory Tracking with Learned Two-Stage Model employing One-Step Future Prediction

Pavlichenko¹,

Behnke²

2021

Preprint

View full text Add to dashboard Cite

Flexible-joint manipulators are frequently used for increased safety during human-robot collaboration and shared workspace tasks. However, joint flexibility significantly reduces the accuracy of motion, especially at high velocities and with inexpensive actuators. In this paper, we present a learningbased approach to identify the unknown dynamics of a flexiblejoint manipulator and improve the trajectory tracking at high velocities. We propose a two-stage model which is composed of a one-step forward dynamics future predictor and an inverse dynamics estimator. The second part is based on linear timeinvariant dynamical operators to approximate the feed-forward joint position and velocity commands. We train the model endto-end on real-world data and evaluate it on the Baxter robot.Our experiments indicate that augmenting the input with onestep future state prediction improves the performance, compared to the same model without prediction. We compare joint position, joint velocity and end-effector position tracking accuracy against the classical baseline controller and several simpler models.

show abstract

Section: Related Workmentioning

confidence: 99%

Flexible-Joint Manipulator Trajectory Tracking with Learned Two-Stage Model employing One-Step Future Prediction

Pavlichenko¹,

Behnke²

2021

Preprint

View full text Add to dashboard Cite

show abstract

LfDT: Learning Dual-Arm Manipulation from Demonstration Translated from a Human and Robotic Arm

Kobayashi,

Yamada,

Hamaya

et al. 2023

2023 IEEE-RAS 22nd International Conference on Humanoid Robots (Humanoids)

View full text Add to dashboard Cite

SimGAN: Hybrid Simulator Identification for Domain Adaptation via Adversarial Reinforcement Learning

Jiang

Zhang

Ho³

et al. 2021

2021 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

As learning-based approaches progress towards automating robot controllers design, transferring learned policies to new domains with different dynamics (e.g. sim-to-real transfer) still demands manual effort. This paper introduces SimGAN, a framework to tackle domain adaptation by identifying a hybrid physics simulator to match the simulated trajectories to the ones from the target domain, using a learned discriminative loss to address the limitations associated with manual loss design. Our hybrid simulator combines neural networks and traditional physics simulaton to balance expressiveness and generalizability, and alleviates the need for a carefully selected parameter set in System ID. Once the hybrid simulator is identified via adversarial reinforcement learning, it can be used to refine policies for the target domain, without the need to collect more data. We show that our approach outperforms multiple strong baselines on six robotic locomotion tasks for domain adaptation.

show abstract

Learning State-Dependent Losses for Inverse Dynamics Learning

Cited by 6 publications

References 16 publications

Flexible-Joint Manipulator Trajectory Tracking with Learned Two-Stage Model employing One-Step Future Prediction

Flexible-Joint Manipulator Trajectory Tracking with Learned Two-Stage Model employing One-Step Future Prediction

LfDT: Learning Dual-Arm Manipulation from Demonstration Translated from a Human and Robotic Arm

SimGAN: Hybrid Simulator Identification for Domain Adaptation via Adversarial Reinforcement Learning

Contact Info

Product

Resources

About