No abstract

No abstract

Recent works in end-to-end control for autonomous driving have investigated the use of vision-based exteroceptive perception. Inspired by such results, we propose a new end-to-end memory-based neural architecture for robot steering and throttle control. We describe and compare this architecture with previous approaches using fundamental error metrics (MAE, MSE) and several external metrics based on their performance on simulated test circuits. The presented work demonstrates the advantages of using internal memory for better generalization capabilities of the model and allowing it to drive in a broader amount of circuits/situations. We analyze the algorithm in a wide range of environments and conclude that the proposed pipeline is robust to varying camera configurations. All the present work, including datasets, network models architectures, weights, simulator, and comparison software, is open source and easy to replicate and extend. Code: github.com/JdeRobot/DeepLearningStudio.

Modular and reconfigurable robotic systems have been designed to provide a customized solution for the non-repetitive tasks to be performed in a constrained environment. Customized solutions are normally extracted from task-based optimization of the possible manipulator configurations but the solution are not integrated, for providing the modular compositions directly. In this work, in the first phase, a strategy of finding unconventional optimal configurations with minimal number of degrees-of-freedom are discussed based upon the prescribed working locations and the cluttered environment. Then, in the second phase, design of the modular and reconfigurable architecture is presented which can adapt these unconventional robotic parameters. Rather than generating and evolving the modular compositions, a strategy is presented through which the unconventional optimal configurations can be mapped directly to the modular compositions. The generated modular composition is validated using Robot Operating System for the motion planning between the prescribed working locations in a given cluttered environment.