Roboter

Kinova Gen3 is a 7DOF torque-controlled robot designed for a variety of robot applications. The robot has seven actuators with integrated torque sensors, a control PC integrated into the first robot’s segment and capable of maintaining 1 kHz closed-loop control, an optional integrated 2D/3D vision module (OmniVision OV5640 color sensor and an Intel RealSense D410 stereo-depth sensor) on the robot’s end-effector, as well as a modular end-effector interface (our current setup uses a Robotiq 2F-85 gripper). The arm can be programmed with a publicly available API (in both Python and C++).

Lego is usually associated with children’s toys, but one can do more than just playing with the Lego NXT robots. Due to the flexible design options, complex systems can be created with it. With the large number of available sensors and actuators, many robotic applications can be realized on such a platform, e.g. navigation and localization. We primarily use the NXTs in the course Scientific Experimentation and Evaluation, where students of the Master of Autonomous Systems program design and perform simple experiments for testing various hypotheses and analyse the experimental results.

Lucy, a Toyota HSR, is meant to be an assistive domestic robot that combines a compact design (the height of the robot is 135cm) with a fairly rich set of sensing and actuation capabilities. Lucy is equipped with an omnidirectional base. The robot’s head, which is a controllable joint as well, has multiple sensors: an RGB-D camera, a wide angle camera, a stereo camera, and a microphone array. Lucy’s manipulator has five degrees of freedom, one of which is a linear joint that can extend the robot’s height. Software-wise, the platform is running ROS.

The youBot is based on an omni-directional platform with four Swedish wheels, a 5 DoF manipulator and a two finger gripper. The robot has been designed to work in industrial-like environments and to perform various tasks as e.g. fetch & carry or assembly tasks. The standard configuration of the youBot has been extended by two Hokuyo laser range finders (one in the front and one in the back), an ASUS Xtion RGB-D camera, Microsoft Lifecam 720p, and a high performance computer (Intel i5 dual core CPU with 2.5 GHz and 2 GB RAM).

QTrobot is an educational social robot developed by LuxAI. The robot’s base is static, but the robot can easily be transported around due to its small weight of about 5kg and height of 64cm. The robot has actuated arms and an actuated neck. QTrobot has an Intel RealSense D435 camera and four microphones; speech output is provided using Acapela Text to Speech; output can also be provided via a screen on the head, which displays a face. QTrobot is integrated with two tablets. Robot applications can be written using ROS; tablet applications are developed using a graphical programming interface.

The NAO robot from Aldebaran joined our robot family back in 2009. The robot comes with a height of 57 cm and a total weight of 5 kg. Internally, the robot is equipped with an Intel Atom CPU, two RGB cameras, a microphone and overall 21 DoF, which allows flexible and comprehensive motions. With its internal battery, NAO is able to operate up to 90 minutes independently of any external power supply.

Lego is usually associated with children’s toys, but one can do more than just playing with the Lego NXT robots. Due to the flexible design options, complex systems can be created with it. With the large number of available sensors and actuators, many robotic applications can be realized on such a platform, e.g. navigation and localization. We primarily use the NXTs in the course Scientific Experimentation and Evaluation, where students of the Master of Autonomous Systems program design and perform simple experiments for testing various hypotheses and analyse the experimental results.

Jenny is based on an omni-wheeled mobile platform called Care-O-Bot 3, developed by the Fraunhofer Institute for Manufacturing Engineering and Automation (IPA). The robot is equipped with a 7 DoF manipulator, a three-fingered hand and an omni-directional platform. The side-mounted tray can be flipped up to the front in order to carry multiple objects simultaneously. The sensor head contains a Stemmer CVC EH6500 HD GE/POE color camera, ASUS Xtion RGB-D camera and a Logitech Microphone. Two SICK S300 laser scanners and one Hokuyo URG-04LX laser scanner are used for mapping and navigation.

Johnny is based on a modular mobile platform called VolksBot which has been designed specifically for rapid prototyping and robot applications in education, research and industry. The VolksBot system is developed by the Fraunhofer Institute for Intelligent Analysis and Information Systems (IAIS). The hardware includes a 5 DoF Katana manipulator with a two finger gripper, a SICK LMS 200 laser range finder in the front for navigation purpose, a Logitech microphone, and a camera head consisting of a PTU-46 and ASUS Xtion RGB-D camera for object detection and recognition.

Bonnie is a logistics robot developed in the ROPOD project. The robot has an omnidirectional base made of so-called smart wheels, which have also been developed as part of the project. A Hokuyo laser is used for autonomous navigation and obstacle avoidance.

Zeno, a Hanson Robokind, is a humanoid robot with a childlike appearance. The robot is about 70 cm tall and can walk upright. With its 11 DoF in its head, Zeno can perform various human-like emotions using facial mimics. An additional microphone and two cameras (in the eyes) are used to perceive the nearby environment.

One of the first robots of our group was the Pioneer 3 mobile platform. Although it is only a four-wheeled mobile platform, it can be equipped with any accessory equipment, such as laser range finders, cameras, torsos and PTUs, or even a manipulator like the 5 DoF Katana manipulator. Due to its robust design, the Pioneer can not only be used indoors, but also in the rough terrain of outdoor environments.