This letter presents a novel cascade state estimation framework for the three-dimensional (3-D) center of mass (CoM) estimation of walking humanoid robots. The proposed framework, called State Estimation RObot Walking (SEROW), fuses effectively joint encoder, inertial, feet pressure, and visual odometry measurements. Initially, we consider the humanoid's Newton-Euler dynamics and rigorously derive the nonlinear CoM estimator. The latter accurately estimates the 3D-CoM position, velocity, and external forces acting on the CoM, while directly considering the presence of uneven terrain and the body's angular momentum rate and, thus, effectively coupling the frontal with the lateral plane dynamics. Furthermore, we extend an established floating mass estimator to take into account the support foot pose, yielding in such a way the mandatory, for CoM estimation, affine transformations and forming a cascade state estimation scheme. Subsequently, we quantitatively and qualitatively assess the proposed scheme by comparing it to other estimation structures in terms of accuracy and robustness to disturbances, both in simulation and on an actual NAO robot walking outdoors over an inclined terrain. To facilitate further research endeavors, our implementation is offered as an open-source ROS/C++ package.

The sense of time is an essential capacity of humans, with a major role in many of the cognitive processes expressed in our daily lifes. So far, in cognitive science and robotics research, mental capacities have been investigated in a theoretical and modelling framework that largely neglects the flow of time. Only recently there has been a small but constantly increasing interest in the temporal aspects of cognition, integrating time into a range of different models of perceptuo-motor capacities. The current paper aims to review existing works in the field and suggest directions for fruitful future work. This is particularly important for the newly developed field of artificial temporal cognition that is expected to significantly contribute in the development of sophisticated artificial agents seamlessly integrated into human societies.

Visual Simultaneous Localization and Mapping (vSLAM)is the method of employing an optical sensor to map the robot’s observable surroundings while also identifying the robot’s pose in relation to that map. The accuracy and speed of vSLAM calculations can have a very significant impact on the performance and effectiveness of subsequent tasks that need to be executed by the robot, making it a key building component for current robotic designs. The application of vSLAM in the area of humanoid robotics is particularly difficult due to the robot’s unsteady locomotion. This paper introduces a pose graph optimization module based on RGB (ORB) features, as an extension of the KinectFusion pipeline (a well-known vSLAM algorithm), to assist in recovering the robot’s stance during unstable gait patterns when the KinectFusion tracking system fails. We develop and test a wide range of embedded MPSoC FPGA designs, and we investigate numerous architectural improvements, both precise and approximation, to study their impact on performance and accuracy. Extensive design space exploration reveals that properly designed approximations, which exploit domain knowledge and efficient management of CPU and FPGA fabric resources, enable real-time vSLAM at more than 30 fps in humanoid robots with high energy-efficiency and without compromising robot tracking and map construction. This is the first FPGA design to achieve robust, real-time dense SLAM operation targeting specifically humanoid robots. An open source release of our implementations and data can be found in [1].