Search

This work investigates the validity of an occupancy grid mapping inspired by human cognition and the way humans visually perceive the environment. This query is motivated by the fact that, to date, no autonomous driving system reaches the performance of an ordinary human driver. The mechanisms behind human perception could provide cues on how to improve common techniques employed in autonomous navigation - specifically the use of occupancy grids to represent the environment. We experiment with a neural network that maps an image of the scene onto an occupancy grid representation, and we show how the model benefits from two key (and yet simple) changes: 1) a different format of occupancy grid that resembles the way the brain projects the environment into a warped representation in the cortical visual area; 2) a mechanism similar to human visual attention that filters out non-relevant information from the scene. These effective expedients can potentially be applied to any autonomous driving task requiring an abstract representation of the scenario like the occupancy grids., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Intelligent Vehicles

This paper presents a novel approach to learning predictive motor control via mental simulations. The method, inspired by learning via mental imagery in natural Cognition, develops in two phases: first, the learning of predictive models based on data recorded in the interaction with the environment; then, at a deferred time, the synthesis of inverse models via offline episodic simulations. Parallelism with human-engineered control-theoretic workflow (mathematical modeling the direct dynamics followed by optimal control inversion) is established. Compared to the latter human-directed synthesis, the mental simulation approach increases autonomy: a robotic agent can learn predictive models and synthesize inverse ones with a large degree of independence. Human modeling is still needed but limited to providing efficient templates for the forward and inverse neural networks and a few other directives. One could consider these templates as the efficient brain network typologies that evolution produced to permit live beings quickly and efficiently learning. The structure of the neural networks both forward and inverse ones; is made of interpretable local models which follows the cerebellar organization (and are also similar to local model approaches known in the literature). We demonstrate the learning of a first-round model (contrasted to Model Predictive Control) for lateral vehicle dynamics. Then, we demonstrate a second learning iteration, where the forward/inverse neural models are significantly improved., CC BY 4.0

In this work an artificial driving agent able to adapt its behavior depending on the specific situation and to generate human-like maneuvers was developed. The agent, called co-driver, was designed with a bio-inspired architecture, with an approach that takes advantage of ideas from cognitive science. In fact the decision process is based on the affordance competition hypothesis: the agent generates a representation of all possible actions given what is perceived in the environment and then chooses the optimal maneuver after inhibiting all the dangerous ones. In the first stage of the development process, atomic actions, called motor primitives, were identified. Then the co-driver was implemented with a layered architecture, where these motor primitives were combined in the upper levels in order to obtain more complex behaviors. Thanks to this particular architecture, the development process of the final artificial driver was split in two parts: initially an advanced driver assistance system (ADAS) was implemented and tested, and finally the architecture was extended in order to have an artificial driving agent for automated driving. In the first case the system was successfully tested on public roads and was able to warn the driver in case of dangerous scenarios, such as blind intersections. In the latter case, the co-driver could manage several challenging test cases in urban scenarios, from car following to cut-in scenario and curves. Finally, it was compared with a state-of-art driver model of CarMaker simulation software used by the main car manufacturers. Due to the achieved results, this work can be considered a valid potential solution for automated driving. In fact, the artificial driver was used as starting point for a European project (“Dreams4Cars”) that is currently in progress.

In this work we propose a number of relevant engineering applications that exhibit both a continuous and a discrete evolution, and are therefore suitably described by a recent formalism for hybrid dynamical systems. More specifically, (i) we design observer schemes for a nonsmooth disturbance arising in AC/DC conversion, which we then cancel from a desirable signal; (ii) we show how reset actuation applied to nonlinear mechanical systems can at the same time sustain or damp oscillations; (iii) we study the feedback interconnection of a classical proportional-integral-derivative controller with a sliding mass under Coulomb friction through differential inclusions. In the context of dynamical systems, we analyze the properties of these applications in terms of asymptotic stability through Lyapunov functions tailored for hybrid systems. Instead of the standard Lyapunov conditions, we prove asymptotic stability through weaker, or relaxed, conditions that are compensated by additional (structural) properties that may be easier to verify.

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.