SFA-MDEN: Semantic-Feature-Aided Monocular Depth Estimation Network Using Dual Branches.

Monocular depth estimation based on unsupervised learning has attracted great attention due to the rising demand for lightweight monocular vision sensors. Inspired by multi-task learning, semantic information has been used to improve the monocular depth estimation models. However, multi-task learning is still limited by multi-type annotations. As far as we know, there are scarcely any large public datasets that provide all the necessary information. Therefore, we propose a novel network architecture Semantic-Feature-Aided Monocular Depth Estimation Network (SFA-MDEN) to extract multi-resolution depth features and semantic features, which are merged and fed into the decoder, with the goal of predicting depth with the support of semantics. Instead of using loss functions to relate the semantics and depth, the fusion of feature maps for semantics and depth is employed to predict the monocular depth. Therefore, two accessible datasets with similar topics for depth estimation and semantic segmentation can meet the requirements of SFA-MDEN for training sets. We explored the performance of the proposed SFA-MDEN with experiments on different datasets, including KITTI, Make3D, and our own dataset BHDE-v1. The experimental results demonstrate that SFA-MDEN achieves competitive accuracy and generalization capacity compared to state-of-the-art methods.

A wrist-inspired suspended tubercle-type tensegrity joint with variable stiffness capacity.

In complex and unpredictable environments or in situations of human-robot interaction, a soft and flexible robot performs more safely and is more impact resistant compared to a traditional rigid robot. To enable robots to have bionic features (flexibility, compliance and variable stiffness) similar to human joints, structures involving suspended tubercle tensegrity are researched. The suspended tubercle gives the joint compliance and flexibility by isolating two moving parts. The variable stiffness capacity is achieved by changing the internal stress of tensegrity through the simultaneous contraction or relaxation of the driving tendons. A wrist-inspired tensegrity-based bionic joint is proposed as a case study. It has variable stiffness and two rotations with a total of three degrees of freedom. Through theoretical derivation and simulation calculation in the NASA Tensegrity RobotToolkit (NTRT) simulator, the range of motion, stiffness adjustable capacity, and their interaction are studied. A prototype is built and tested under a motion capture system. The experimental result agrees well with the theoretical simulation. Our experiments show that the suspended tubercle-type tensegrity is flexible, the stiffness is adjustable and easy to control, and it has great potential for bionic joints.