Trajectory adjustment for nonprehensile manipulation using latent space of trained sequence-to-sequence model*

When robots are used to manipulate objects in various ways, they often have to consider the dynamic constraint. Machine learning is a good candidate for such complex trajectory planning problems. However, it sometimes does not satisfy the task objectives due to a change in the objective or a lack of guarantee that the objective functions will be satisfied. To overcome this issue, we applied a method of trajectory deformation by using sequence-to-sequence (seq2seq) models. We propose a method of adjusting the generated trajectories, by utilizing the architecture of seq2seq models. The proposed method optimizes the latent variables of the seq2seq models instead of the trajectories to minimize the given objective functions. The verification results show that the use of latent variables can obtain the desired trajectories faster than direct optimization of the trajectories. GRAPHICAL ABSTRACT

[1]  Kevin M. Lynch,et al.  Planning and control for dynamic, nonprehensile, and hybrid manipulation tasks , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[2]  Florent Lamiraux,et al.  Reactive trajectory deformation for nonholonomic systems: application to mobile robots , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[3]  Bowen Zhou,et al.  Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond , 2016, CoNLL.

[4]  Sandra Hirche,et al.  Online deformation of optimal trajectories for constrained nonprehensile manipulation , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[5]  Nolan Wagener,et al.  Learning contact-rich manipulation skills with guided policy search , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[6]  Toshiaki Tsuji,et al.  Optimized Trajectory Generation based on Model Predictive Control for Turning Over Pancakes , 2018 .

[7]  Toshiaki Tsuji,et al.  Model predictive control based deep neural network for dynamic manipulation , 2017, IECON 2017 - 43rd Annual Conference of the IEEE Industrial Electronics Society.

[8]  Zoran Popovic,et al.  Interactive Control of Diverse Complex Characters with Neural Networks , 2015, NIPS.

[9]  Kevin M. Lynch,et al.  Nonprehensile robotic manipulation: controllability and planning , 1996 .

[10]  Quang-Cuong Pham,et al.  Dynamic non-prehensile object transportation , 2014, 2014 13th International Conference on Control Automation Robotics & Vision (ICARCV).

[11]  Sandra Hirche,et al.  Fast trajectory replanning using Laplacian mesh optimization , 2012, 2012 12th International Conference on Control Automation Robotics & Vision (ICARCV).

[12]  Paul Umbanhowar,et al.  Sliding manipulation of rigid bodies on a controlled 6-DoF plate , 2012, Int. J. Robotics Res..

[13]  Sergey Levine,et al.  Self-Consistent Trajectory Autoencoder: Hierarchical Reinforcement Learning with Trajectory Embeddings , 2018, ICML.

[14]  Andrej Gams,et al.  User Feedback in Latent Space Robotic Skill Learning , 2018, 2018 IEEE-RAS 18th International Conference on Humanoid Robots (Humanoids).

[15]  Quoc V. Le,et al.  Sequence to Sequence Learning with Neural Networks , 2014, NIPS.

[16]  Makoto Kaneko,et al.  Dynamic Manipulation Inspired by the Handling of a Pizza Peel , 2009, IEEE Transactions on Robotics.

[17]  Yoshua Bengio,et al.  Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation , 2014, EMNLP.

[18]  Sergey Levine,et al.  Learning deep control policies for autonomous aerial vehicles with MPC-guided policy search , 2015, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[19]  Ahmad A. Masoud,et al.  Kinodynamic Motion Planning , 2010, IEEE Robotics & Automation Magazine.

[20]  Martin Wattenberg,et al.  Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation , 2016, TACL.

[21]  Shigeki Sugano,et al.  Learning to Perceive the World as Probabilistic or Deterministic via Interaction With Others: A Neuro-Robotics Experiment , 2017, IEEE Transactions on Neural Networks and Learning Systems.

[22]  Jason Yosinski,et al.  Deep neural networks are easily fooled: High confidence predictions for unrecognizable images , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[23]  Tetsuya Ogata,et al.  Dynamical Integration of Language and Behavior in a Recurrent Neural Network for Human–Robot Interaction , 2016, Front. Neurorobot..

[24]  Steven M. LaValle,et al.  Randomized Kinodynamic Planning , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).

[25]  Jörg Conradt,et al.  Serendipitous Offline Learning in a Neuromorphic Robot , 2016, Front. Neurorobot..

[26]  Toshiaki Tsuji,et al.  Sequence-to-Sequence Model for Trajectory Planning of Nonprehensile Manipulation Including Contact Model , 2018, IEEE Robotics and Automation Letters.

[27]  Thomas Brox,et al.  Synthesizing the preferred inputs for neurons in neural networks via deep generator networks , 2016, NIPS.

[28]  Toshiaki Tsuji,et al.  Sequence-to-sequence models for trajectory deformation of dynamic manipulation , 2017, IECON 2017 - 43rd Annual Conference of the IEEE Industrial Electronics Society.

[29]  Sergey Levine,et al.  Learning Latent Plans from Play , 2019, CoRL.

[30]  Toshiaki Tsuji,et al.  Dynamic Object Manipulation Considering Contact Condition of Robot With Tool , 2016, IEEE Transactions on Industrial Electronics.

[31]  Samy Bengio,et al.  Generating Sentences from a Continuous Space , 2015, CoNLL.

[32]  Thierry Fraichard,et al.  From path to trajectory deformation , 2007, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[33]  Danica Kragic,et al.  Rearrangement with Nonprehensile Manipulation Using Deep Reinforcement Learning , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[34]  Kouhei Ohnishi,et al.  Motion control for advanced mechatronics , 1996 .