In current decades, significant advancements in robotics engineering and autonomous vehicles have improved the requirement for precise depth measurements. Depth estimation (DE) is a traditional task in computer vision that can be appropriately predicted by applying numerous procedures. This task is vital in disparate applications such as augmented reality and target tracking. Conventional monocular DE (MDE) procedures are based on depth cues for depth prediction. Various deep learning techniques have demonstrated their potential applications in managing and supporting the traditional ill-posed problem. The principal purpose of this paper is to represent a state-of-the-art review of the current developments in MDE based on deep learning techniques. For this goal, this paper tries to highlight the critical points of the state-of-the-art works on MDE from disparate aspects. These aspects include input data shapes and training manners such as supervised, semi-supervised, and unsupervised learning approaches in combination with applying different datasets and evaluation indicators. At last, limitations regarding the accuracy of the DL-based MDE models, computational time requirements, real-time inference, transferability, input images shape and domain adaptation, and generalization are discussed to open new directions for future research.
1 February 2019 in International Journal of Social Robotics
A tremendous amount of research is being performed regarding robot interaction with individuals having intellectual disability, especially for kids with Autism Spectrum Disorders (ASD). These researches have shown many promising advancements about the use of interactive robots for rehabilitation of such individuals. However, these studies fail to analyze and explore the effects of robotics interaction with individuals having profound and multiple learning disabilities (PMLD). This research presents a thorough case study regarding interaction of individuals having PMLD with a humanoid robot in different possible categories of robotic interaction. Separate interaction activities are designed as a representative for the different categories of possible clinical applications of the interactive robot. All the trials were assessed using different evaluation techniques. Finally, the results strongly suggest that robotic interactions can help to induce a target behavior among these individuals, to teach and to encourage them which can bring an autonomy to certain extent in their life.
A closed‐loop system for the central pattern generator (CPG)‐based locomotion control of biped robots that operates in the joint space is presented. The proposed system has been designed to allow biped robots to walk on unknown sloped surfaces. Feedback signals generated by the robot's inertial and force sensors are directly fed into the CPG to automatically adjust the locomotion pattern over flat and sloped terrain in real time. The proposed control system negotiates sloped surfaces similarly to state‐of‐the‐art CPG‐based control systems; however, whereas the latter must continuously solve the computationally intensive inverse kinematics of the robot, the proposed approach directly operates in the joint space, which makes it especially suitable for direct hardware implementation with electronic circuits. The performance of the proposed control system has been assessed through both simulation and real experiments on a NAO humanoid robot.
1 December 2015 in Robotics and Autonomous Systems