|
Design and Grasping Control of Rib-Reinforced Bending Soft Actuators by Vacuum Driven |
|
Ryotaro Taguchi(Kyoto Institute of Technology, Japan) |
| The Rib-Reinforced Vacuum-Driven Actuator was designed to work safely with humans to perform tasks. Most of the soft actuators are powered by compressed air, however, there is a risk of rupture. In this research, a safe vacuum-driven soft actuator has been developed to prevent rupture. In order to prevent undesirable buckling due to atmospheric pressure, Ribs are placed inside to enable bending in vacuum. This soft actuator, which generates a maximum force of about 2.9N at the tip, can force control using the pressure-force characteristics. The soft robot hand can grasp objects up to 0.535 kg. In addition, the force control allowed the soft robot hand to gently grasp the paper balloon. |
|
|
Dynamic Obstacle Avoidance of Multi-rotor UAV using Chance Constrained MPC |
|
Takumi Wakabayashi, Yuma Nunoya, Satoshi Suzuki(Chiba University, Japan) |
| One of the main issues in motion planning among multiple Unmanned Aerial Vehicles is collision avoidance. Chance constrained Model Predictive Control is characterized by its ability to consider collision probabilities in a constrained optimization framework. However, the structure of the collision probability constraint equation to be introduced into the evaluation function has not been sufficiently studied. In this study, the structure of equations for incorporating probability constraints into the evaluation function is examined and compared with methods using deterministic constraints. |
|
|
Vision-Based 3D Reconstruction Using a Compound Eye Camera |
|
Wooseok Oh, Hwiyeon Yoo, Timothy Ha, Songhwai Oh(Seoul National University, Korea) |
| Vision-based 3D reconstruction is an important problem for tasks like naviation. Although various vision-based methods are being studied, it is difficult to reconstruct many parts at once with a general camera because of a small FOV. To solve this problem, we propose a coarse but lightweight reconstruction method using a camera with a unique structure called a compound eye with various advantages such as large FOV. In the process, we devise a network that performs depth estimation on a compound eye structure. We tested our methods in a Gazebo simulation, and it can capture the simulation environment with a high recall of 97.51%. |
|
|
Robust Control of the Aerial Manipulator with a Fixed End-effector Position |
|
Jeonghyun Byun, H. Jin Kim(Seoul National University, Korea) |
| - The necessity for aerial manipulation while grasping a fixed point is on the rise to broaden the range of tasks that can be performed with flying robots such as plug pulling or drawer knob grasping.
- Using the constrained Euler-Lagrange equation, a dynamic equation for the aerial manipulator which is freely rotating around the fixed point is derived.
- A disturbance-observer-based (DOB) control law is constructed.
- A numerical simulation was conducted and the results show that the Euler angles satisfactorily follow their desired trajectory.
- For future works, the proposed controller could be applied to actual experiments. |
|
|
Collision-free Transport of 2D Deformable Objects |
|
Rafael Herguedas, Gonzalo Lopez-Nicolas, Carlos Sagues(Instituto Universitario de Investigación en Ingeniería de Aragón, Spain) |
| We propose a novel system to transport 2D cloth-like deformable objects with mobile manipulators and without collisions along a known path. First, a new deformation model that allows for real-time shape prediction, based on the paradigm of deformable bounding box, is presented. The transport task is next defined as an optimization problem, which includes a set of linear and nonlinear constraints. These constraints allow to limit the object's deformations and rotations and to avoid obstacles, respectively. Simulation results are reported to demonstrate the validity of our method. |
|
|
Semantic Segmentation of Outcrop Images using Deep Learning Networks Toward Realization of Carbon Capture and Storage |
|
Kodai Sato(Akita Prefectural University, Japan), Hirokazu Madokoro(Iwate Prefectural University, Japan), Takeshi Nagayoshi(Akita Prefectural University, Japan), Shun Chiyonobu, Paolo Martizzi(Akita University, Japan), Stephanie Nix(Akita Prefectural University, Japan), Hanwool Woo(The University of Tokyo, Japan), Takashi K. Saito, Kazuhito Sato(Akita Prefectural University, Japan) |
| This study was conducted to classify outcrop images using semantic segmentation methods based on deep learning algorithms. Carbon capture and storage (CCS) is an epoch-making approach to reduce greenhouse gases in the atmosphere. This study specifically examines outcrops because geological layer measurements can lead to production of a highly accurate geological model for feasible CCS inspections. Experimentally obtained results demonstrated that data expansion with random sampling improved the accuracy. |
|