Neuro-Genetic Visuomotor Architecture for Robotic Grasping

Abstract

We present a novel, hybrid neuro-genetic visuomotor architecture for object grasping on a humanoid robot. The approach combines the state-of-the-art object detector RetinaNet, a neural network-based coordinate transformation and a genetic-algorithm-based inverse kinematics solver. We claim that a hybrid neural architecture can utilise the advantages of neural and genetic approaches: while the neural components accurately locate objects in the robot’s three-dimensional reference frame, the genetic algorithm allows reliable motor control for the humanoid, despite its complex kinematics. The modular design enables independent training and evaluation of the components. We show that the additive error of the coordinate transformation and inverse kinematics solver is appropriate for a robotic grasping task. We additionally contribute a novel spatial-oversampling approach for training the neural coordinate transformation that overcomes the known issue of neural networks with extrapolation beyond training data and the extension of the genetic inverse kinematics solver with numerical fine-tuning. The grasping approach was realised and evaluated on the humanoid robot platform NICO in a simulation environment.