Exploiting dream-like simulation mechanisms to develop safer agents for automated driving: The “Dreams4Cars” EU research and innovation action

Automated driving needs unprecedented levels of reliably and safety before marked deployment. The average human driver fatal accident rate is 1 every 100 million miles. Automated vehicles will have to provably best these figures. This paper introduces the notion of dream-like mechanisms as a simulation technology to produce a large number of hypothetical design and test scenarios — especially focusing on variations of more frequent dangerous and near miss events. Grounded in the simulation hypothesis of cognition, we show here some principles for effective simulation mechanisms and an artificial cognitive system architecture that can learn from the simulated situations.

[1]  Andrea Vedaldi,et al.  Understanding deep image representations by inverting them , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  E. Bertolazzi,et al.  Symbolic–Numeric Indirect Method for Solving Optimal Control Problems for Large Multibody Systems , 2005 .

[3]  David Windridge,et al.  Perception-action learning as an epistemologically-consistent model for self-updating cognitive representation. , 2010, Advances in experimental medicine and biology.

[4]  Emanuel Todorov,et al.  Evidence for the Flexible Sensorimotor Strategies Predicted by Optimal Feedback Control , 2007, The Journal of Neuroscience.

[5]  Thomas Brox,et al.  Learning to Generate Chairs, Tables and Cars with Convolutional Networks , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  T. Flash,et al.  Minimum-jerk, two-thirds power law, and isochrony: converging approaches to movement planning. , 1995, Journal of experimental psychology. Human perception and performance.

[7]  Damian Harty,et al.  The Multibody Systems Approach to Vehicle Dynamics , 2004 .

[8]  Aude Billard,et al.  Action Selection and Mental Transformation Based on a Chain of Forward Models , 2004 .

[9]  Ashvin Shah,et al.  Finding minimal action sequences with a simple evaluation of actions , 2014, Front. Comput. Neurosci..

[10]  Arend L. Schwab,et al.  Dynamics of Multibody Systems , 2007 .

[11]  Francesco Biral,et al.  An intelligent curve warning system for powered two wheel vehicles , 2010 .

[12]  Sethu Vijayakumar,et al.  Adaptive Optimal Feedback Control with Learned Internal Dynamics Models , 2010, From Motor Learning to Interaction Learning in Robots.

[13]  Daniel M. Wolpert,et al.  The Main Sequence of Saccades Optimizes Speed-accuracy Trade-off , 2006, Biological Cybernetics.

[14]  E. Bertolazzi,et al.  real-time motion planning for multibody systems , 2007 .

[15]  Bartlett W. Mel.  A Connectionist Model May Shed Light on Neural Mechanisms for Visually Guided Reaching , 1991, Journal of Cognitive Neuroscience.

[16]  Ron Chrisley,et al.  Cognitive Map Construction and Use: A Parallel Distributed Processing Approach , 1991 .

[17]  Germund Hesslow,et al.  EXPLORING INTERNAL SIMULATION OF PERCEPTION IN MOBILE ROBOTS , 2001 .

[18]  David Windridge,et al.  A linear-complexity reparameterisation strategy for the hierarchical bootstrapping of capabilities within perception-action architectures , 2009, Image Vis. Comput..

[19]  Thomas Brox,et al.  Inverting Visual Representations with Convolutional Networks , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  G. Hesslow.  The current status of the simulation theory of cognition , 2012, Brain Research.

[21]  Kevin N. Gurney,et al.  The Basal Ganglia and Cortex Implement Optimal Decision Making Between Alternative Actions , 2007, Neural Computation.

[22]  Tom Ziemke,et al.  Internal simulation of perception: a minimal neuro-robotic model , 2005, Neurocomputing.

[23]  Sean R. Anderson,et al.  Adaptive filters and internal models: Multilevel description of cerebellar function , 2013, Neural Networks.

[24]  Daniel A. Braun,et al.  Optimal Control Predicts Human Performance on Objects with Internal Degrees of Freedom , 2009, PLoS Comput. Biol..

[25]  E. Bertolazzi,et al.  1 SYMBOLIC-NUMERIC INDIRECT METHOD FOR SOLVING OPTIMAL CONTROL PROBLEMS FOR LARGE MULTIBODY SYSTEMS : THE TIME-OPTIMAL RACING VEHICLE EXAMPLE , 2003 .

[26]  David Windridge,et al.  Artificial Co-Drivers as a Universal Enabling Technology for Future Intelligent Vehicles and Transportation Systems , 2015, IEEE Transactions on Intelligent Transportation Systems.

[27]  Abhinandan Jain.  Robot and Multibody Dynamics: Analysis and Algorithms , 2010 .

[28]  Horst-Michael Groß,et al.  Generative character of perception: a neural architecture for sensorimotor anticipation , 1999, Neural Networks.

[29]  E. Todorov.  Optimality principles in sensorimotor control , 2004, Nature Neuroscience.

[30]  Huei Peng,et al.  Vehicle dynamics applications of optimal control theory , 2011 .

[31]  Stefano Nolfi,et al.  Extracting Regularities in Space and Time Through a Cascade of Prediction Networks: The Case of a Mobile Robot Navigating in a Structured Environment , 1999, Connect. Sci..

[32]  Tom Ziemke,et al.  Finding Your Way from the Bed to the Kitchen: Reenacting and Recombining Sensorimotor Episodes Learned from Human Demonstration , 2016, Front. Robot. AI.

[33]  Francesco Biral,et al.  Supporting Drivers in Keeping Safe Speed and Safe Distance: The SASPENCE Subproject Within the European Framework Programme 6 Integrating Project PReVENT , 2010, IEEE Transactions on Intelligent Transportation Systems.

[34]  Kevin Gurney,et al.  Biologically Guided Driver Modeling: the Stop Behavior of Human Car Drivers , 2018, IEEE Transactions on Intelligent Transportation Systems.

[35]  Paul Cisek,et al.  Cortical mechanisms of action selection: the affordance competition hypothesis , 2007, Philosophical Transactions of the Royal Society B: Biological Sciences.

[36]  Mauro Da Lio,et al.  Cooperative Intersection Support System Based on Mirroring Mechanisms Enacted by Bio-Inspired Layered Control Architecture , 2018, IEEE Transactions on Intelligent Transportation Systems.

[37]  Tom Ziemke,et al.  Dreaming of electric sheep? Exploring the functions of dream-like mechanisms in the development of mental imagery simulations , 2013, Adapt. Behav..

[38]  A Framework for Hierarchical Perception–Action Learning Utilizing Fuzzy Reasoning , 2013, IEEE Transactions on Cybernetics.

[39]  Roberto Lot,et al.  A Symbolic Approach for Automatic Generation of the Equations of Motion of Multibody Systems , 2004 .

[40]  Christoph Adami,et al.  What Do Robots Dream Of? , 2006, Science.

[41]  Stefano Nolfi,et al.  Learning to perceive the world as articulated: an approach for hierarchical learning in sensory-motor systems , 1998, Neural Networks.

[42]  Kevin N. Gurney,et al.  A biologically plausible embodied model of action discovery , 2012, Front. Neurorobot..