Multi-Abstractive Neural Controller: An Efficient Hierarchical Control Architecture for Interactive Driving

As learning-based methods make their way from perception systems to planning/control stacks, robot control systems have started to enjoy the benefits that data-driven methods provide. Because control systems directly affect the motion of the robot, data-driven methods, especially black box approaches, need to be used with caution considering aspects such as stability and interpretability. In this letter, we describe a differentiable and hierarchical control architecture. The proposed representation, called multi-abstractive neural controller, uses the input image to control the transitions within a novel discrete behavior planner (referred to as the visual automaton generative network, or vAGN). The output of a vAGN controls the parameters of a set of dynamic movement primitives which provides the system controls. We train this neural controller with real-world driving data via behavior cloning and show improved explainability, sample efficiency, and similarity to human driving.

[1]  Byron Boots,et al.  Learning to Optimize in Model Predictive Control , 2022, 2022 International Conference on Robotics and Automation (ICRA).

[2]  Danilo Jimenez Rezende,et al.  Learning to Induce Causal Structure , 2022, ICLR.

[3]  D. Scaramuzza,et al.  Real-Time Neural MPC: Deep Learning Model Predictive Control for Quadrotors and Agile Robotic Platforms , 2022, IEEE Robotics and Automation Letters.

[4]  Sicun Gao,et al.  Safe Control With Learned Certificates: A Survey of Neural Lyapunov, Barrier, and Contraction Methods for Robotics and Control , 2022, IEEE Transactions on Robotics.

[5]  S. Levine,et al.  RvS: What is Essential for Offline RL via Supervised Learning? , 2021, ICLR.

[6]  Francesco Borrelli,et al.  Accelerating Quadratic Optimization with Reinforcement Learning , 2021, NeurIPS.

[7]  Seong Joon Oh,et al.  Neural Hybrid Automata: Learning Dynamics with Multiple Modes and Stochastic Transitions , 2021, NeurIPS.

[8]  Vineel Pratap,et al.  Differentiable Weighted Finite-State Transducers , 2020, ArXiv.

[9]  Radu Grosu,et al.  Neural circuit policies enabling auditable autonomy , 2020, Nature Machine Intelligence.

[10]  Sebastian Starke,et al.  Neural state machine for character-scene interactions , 2019, ACM Trans. Graph..

[11]  Ken Goldberg,et al.  Multi-Task Hierarchical Imitation Learning for Home Automation , 2019, 2019 IEEE 15th International Conference on Automation Science and Engineering (CASE).

[12]  Christopher D. Manning,et al.  Learning by Abstraction: The Neural State Machine , 2019, NeurIPS.

[13]  Alexander Liniger,et al.  Learning-Based Model Predictive Control for Autonomous Racing , 2019, IEEE Robotics and Automation Letters.

[14]  Sergio Casas,et al.  End-To-End Interpretable Neural Motion Planner , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  Dieter Fox,et al.  Neural Autonomous Navigation with Riemannian Motion Policy , 2019, 2019 International Conference on Robotics and Automation (ICRA).

[16]  Qiang Xu,et al.  nuScenes: A Multimodal Dataset for Autonomous Driving , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Wael Farag,et al.  Behavior Cloning for Autonomous Driving using Convolutional Neural Networks , 2018, 2018 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT).

[18]  Henggang Cui,et al.  Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks , 2018, 2019 International Conference on Robotics and Automation (ICRA).

[19]  S. Gershman,et al.  How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanation , 2018, ArXiv.

[20]  Lydia Tapia,et al.  PRM-RL: Long-range Robotic Navigation Tasks by Combining Reinforcement Learning and Sampling-Based Planning , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[21]  Mariusz Bojarski,et al.  VisualBackProp: visualizing CNNs for autonomous driving , 2016, ArXiv.

[22]  Zachary Chase Lipton.  The mythos of model interpretability , 2016, ACM Queue.

[23]  Bolei Zhou,et al.  Learning Deep Features for Discriminative Localization , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Jimmy A. Jørgensen,et al.  Adaptation of manipulation skills in physical contact with the environment to reference force profiles , 2015, Auton. Robots.

[25]  Qiang Chen,et al.  Network In Network , 2013, ICLR.

[26]  S. Schaal,et al.  Dynamical Movement Primitives: Learning Attractor Models for Motor Behaviors , 2013, Neural Computation.

[27]  Jun Nakanishi,et al.  Movement imitation with nonlinear dynamical systems in humanoid robots , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[28]  L. Kochiev,et al.  Neural State Machine for 2D and 3D Visual Question Answering , 2021 .

[29]  Zoe Doulgeri,et al.  A correct formulation for the Orientation Dynamic Movement Primitives for robot control in the Cartesian space , 2019, CoRL.