Intelligent Control and Robotic Systems Lab

This paper presents the SMARoll robot which is a modular closed-chain rolling robot with compliant shape memory alloy (SMA) actuators that can mimic the behavior of rolling animals. The SMARoll robot can provide advanced mobility, controllability, terrain adaptability and modularity/scalability. It also has the capability to overcome obstacles and snags that might be useful to explore unstructured/rugged environments. The robot consists of twelve links and joints. It is actuated by a novel SMA rotating actuator at each joint which provides compliance/flexibility advantage to the robot. The rotating actuator is fabricated in a novel manner by sewing a SMA wire between two rotating frames. Kinematic analysis for the SMARoll robot is studied to determine the configuration of the entire robot considering links position and orientation as a function of joint angles. At an equilibrium position, the SMARoll robot is mostly resting on four links. To move the robots center of mass and the whole robot, the locomotion of the robot can be divided into steps. At each step, only four (2 active and 2 passive) joint motions are required for this robot locomotion strategy. Experimental work is performed to validate the robot design and kinematic analysis and prove locomotion strategy. The experimental results show that the robot has a good capability to achieve terrain adaptability due to compliance advantage. Also, results show that the robot is capable to achieve a speed of 0.13 m/s at plain ground. (C) 2020 Elsevier B.V. All rights reserved.

In this work, novel Shape Memory Alloy (SMA)-based actuators are proposed to provide angular displacements in both clockwise and counter-clockwise directions with compliance. A bidirectional SMA rotating actuator is fabricated using a rotating frame and two SMA wire-based actuating units similar to human skeletal muscle systems without any additional complicated rotational driving mechanism. These actuating units are activated independently to provide bidirectional rotary motions by using sequentially coordinated electrical inputs. The mechanical, thermal, and electrical properties of the bidirectional SMA rotating actuator are also characterized experimentally. The design and manipulation of the proposed actuator are experimentally verified with simple open-loop and closed-loop control strategies. (C) 2019 Elsevier B.V. All rights reserved.

In this research a biologically inspired finger-like mechanism similar to human musculoskeletal system is developed based on Shape Memory Alloys (SMAs). SMA actuators are inspiring the design of a modular finger part with compact and compliant actuation. A three-segmented finger-like mechanism is designed and fabricated. This mechanism is composed of six bending Shape Memory Alloy (SMA) actuators. As a result, our finger mechanism is compact and compliant. The insider three SMA actuators are used for finger flexion while the outsider three SMA actuators are for extension. Each segment of this mechanism can be bent and/or extended independently by actuating a corresponding bending SMA actuator. Furthermore, full bending motion can be achieved by applying coordinated control of the three SMA actuators. Toward this goal a mathematical model of the SMA combined finger has been developed. The developed mathematical model is then used to design a proportional-derivative controller for control compliant actuation of the finger-mechanism. The performance of this mechanism has been experimentally evaluated. Our experimental results verify that the SMA-based finger module can achieve the desired postures similar to a human finger.

This paper presents a novel miniaturized and modular dual-axis Electromagnetic Actuator (EMA). It mainly consists of two electromagnetic coils in an orthogonal orientation with a permanent magnet fixed on a free moving frame that rotates around two axes/joints. By actuating either of the coils, the free moving frame rotates around the corresponding axis. Simulations and experimental analyses are conducted in order to characterize the performance of our EMA. Thus, our actuator achieves a torque of 100 mNm at simulation and 80 mNm through experimentation for the same applied current. Additionally, it can achieve a rotation of 10 degrees (approximate to 0.2 rad), according to simulations and experimental work. Because of modularity, multiple units of our EMA can be connected together in different configuration to serve in several applications. As an example application, we used a pair of our EMA in order to generate a miniaturized 4-DOF robotic manipulator. This manipulator demonstrates the advantages of light weight, small size, and a high level of manipulability. Kinematic analyses and experimental work are performed in order to validate our manipulator and to prove the concept of our proposed EMA. Through this experiment, we applied an open-loop controller on our EMAs, so that the end-effector of our manipulator can track a predefined circular trajectory. The movement of the end-effector is detected while using image processing techniques. Although we used an open-loop controller, our manipulator is still able to track the trajectory with moderate errors.

In this research we present an algorithm for a six-wheeled robotic vehicle with articulated suspension (RVAS) to estimate the vehicle velocity and acceleration states, slip ratio and the tire forces. The estimation algorithm consists of six parts. In the first part, a wheel state estimator estimates the wheel rotational speed and its angular acceleration using Kalman filter, which is used to estimate the longitudinal tire force distribution in the second part. The third part is to estimate respective longitudinal, lateral, and vertical speeds of the vehicle and wheels. Based on these speeds, the slip ratio and slip angle are estimated in the fourth part. In the fifth part, the vertical tire force is then estimated. In the sixth part, the lateral tire force is then estimated. For a simulation test environment, the RVAS dynamic model is developed using Matlab and Simulink. The estimation algorithm is then verified in simulation using the vehicle test data and different test scenarios. It is found from simulation results that the proposed estimation algorithm can estimate the vehicle states, longitudinal tire forces efficiently. Moreover, a small prototype of the robotic vehicle is fabricated for experimental verification of the estimation algorithm. Various experiments are executed in pavement and off-road driving to estimate the wheel angular position, velocity and acceleration states and finally the slip ratio is estimated in these situations.

In this study, a six-wheeled small mobile robot platform for rough terrain is developed. Speeds, accelerations, and slip ratios of the robot are estimated. We analyzed a vehicle with an articulated suspension (RVAS: Robot vehicle with articulated suspension) for actual rough terrain. We then design a small RVAS mobile robot by reducing the scale. The mobile robot is modeled and fabricated by using 3D printing technique. Considering physical limitations of small mobile robots and available sensors, one microcontroller, six hall sensor encoders, and one IMU sensor are used. A Kalman filter-based estimator is also used to remove noise from the acquired data and improve estimation of the robot states. The slip rate of each wheel is experimentally estimated by using the angular velocity of each wheel and the linear velocity of the robot acquired from the IMU sensor.

In this study, we designed and fabricated a transformable wheel-leg robot that provides efficient mobility and obstacle overcoming under various structured and unstructured environments. The transformable wheel-leg is configured as a circular wheel on a flat surface and into the leg mode on rugged terrain for advanced mobility and terrain adaptability. Based on this goal, we analyzed the design parameters of the wheel-leg. A DC motor and gear mechanism were applied to realize transformation of the wheel and leg modes in the robot. Finally, experiments were performed to verify the improved obstacle overcoming capability.

오늘날 무인기는 군사, 기계 , 항공우주공학분야에서 정보수집, 정찰비행 및 수색작전 등을 위해 널리 사용된다. 무인기 구동제어기는 비행 시 발생하는 부하의 영향을 받게 되며, 이러한 부하상태에서 정밀한 위치제어가 실현되어야 한다. 구동 제어기가 사용되는 일반적인 주위의 환경 온도는 약 -30C-50C(243K-323K)이다. 구동 제어기가 신뢰성을 갖추기 위해 중요시 되는 것은 구동 제어기 내의 열 발생이 어느 한계 수준(-25C-105C(243K-378K)내로 유지되어야 한다. 따라서 구동제어기 내의 열유동 특성에 대한 연구와 분석이 필요하게 된다. 본 논문에서는 3-D 모델링을 위해 상용 소프트웨어인 Pro/E와 Solid-Works를, 수치 시뮬레이션을 위해서는 Cosmos/Flow Works를 사용하였다. 본 논문에서는 내부의 칩들과 보드들을 지니는 구동 제어기의 열유동 특성을 통하여 안전한 구동 제어기의 설계를 제시한다.

The 2-DOF IPMC module is introduced in this research. Each rotation axis of 2-DOF IPMC module is perpendicular each other. A second segment of IPMC is attached across to end tap of the first segment of IPMC directly without any frame. The link method with 2 IPMC actuator is assemble using a slot simply and conductive epoxy adhesive is used for wiring. The single IPMC strips are connected to rigid links, the open-loop control of the IPMC manipulator. The inverse kinematics of the linked 2-DOF IPMC was established. The experimental results show 2-DOF motion of the IPMC actuator module. This 2-DOF IPMC actuator module can be improved or modified for example by adding more links to even more increase the workspace by adding an extra soft link to the tip of the manipulator for extra soft manipulation.

The aim of this study is to investigate the turning locomotion of a worm-like robot using four segmented solenoids. The electromagnetic force driven worm-like robot is composed of four solenoids, a compliance guide, and a tendon-driven mechanism. The structure of the actuator in the robot mimics the muscular structure and locomotion of an earthworm. The turning motion of the robot has been analyzed both theoretically and numerically. Electromagnetic force and friction force govern the motion of the robot. The prototype of the robot was fabricated based on the simulation study, and its performance was compared to the simulation study. The robot is able to turn along a guide of 1° with a friction of 0.5 and 500 A-turn. It was demonstrated that the worm-like robot achieved turning locomotion.

본 연구에서는 제한된 H/W 성능을 가진 소형 모바일 로봇의 성공적인 자율주행을 위하여 인공지능을 적용하여 주변의 환경을 효과적으로 인식하기 위한 연구를 수행하였다. 이를 위하여 먼저 비정형 환경에서 구동 바퀴의 형태를 자유롭게 변형하여 험지를 효과적으로 극복할 수 있는 소형 휠-레그(wheel-leg) 모바일 로봇을 설계하였다. 험지를 효율적으로 주행하기 위해 로봇의 바퀴에 기어구조를 적용하여 휠-레그의 크기가 최대 2.5배까지 변형할 수 있도록 설계하였다. 주변 환경의 인식을 위해 GPU가 내장된 저가형 임베디드 보드와 카메라를 사용하였다. 본 연구에서 사용한 딥러닝 모델은 YOLOv3를 사용하였고 학습한 모델의 성능을 mAP, GIoU 등으로 확인하였다. 본 연구를 통하여 제안된 객체 인식 알고리즘을 실제 로봇에 적용하여 로봇이 비정형 환경에서 효과적으로 장애물을 인식하고 이를 극복하는 것을 확인하였다.