ROS is an open-source, meta-operating system for robots, developed by Willow Garage. It provides the services you would expect from an operating system, including hardware abstraction, low-level device control, implementation of commonly-used functionality, message-passing between processes, and package management.

Player is a device server that provides a powerful, flexible interface to a variety of sensors and actuators (e.g., robots). Because Player uses a client/server model, robot control programs can be written in any programming language and can execute on any computer with network connectivity to the robot. In addition, Player supports multiple concurrent client connections to devices, creating new possibilities for distributed and collaborative sensing and control.

Stage is a scaleable multiple robot simulator; it simulates a population of mobile robots moving in and sensing a two- dimensional bitmapped environment, controlled through Player. Stage provides virtual Player robots which interact with simulated rather than physical devices. Various sensor models are provided, including sonar, scanning laser rangefinder, pan-tilt-zoom camera with color blob detection and odometry.

Gazebo complements stage's simulation capabilities, focusing on higher fidelity dynamics based simuation in a 3D environment. ROS or Player is still used as an interface to the robot, making simulations in Gazebo, Stage and physical robots an easy task to accomplish.

More information on the Player/Stage/Gazebo suite of software is available at this project page. It also has some links to the official Sourceforge.net project page.

The SARCOS Humanoid is a hydraulic full-body humanoid robot with 44 degrees of freedom, an active vision system with two moving eyes, actuated hands, an inbuilt Kinect 3D camera, and various touch and IMU sensing devices. The robot is part of the USC Human Centered Robotics lab for research in robot control, autonomous learning, and human robot interaction.

The NAO is a full-body humanoid robot manufactured by Aldebaran Robtoics. It is approximately 26 in (66 cm) tall with 26 degrees of freedom in the neck, shoulders, elbows, hands, hips, knees, wrists, and ankles. It has a number of sensors, including an IMU, sonar, video, microphones, and force sensors. It has two speakers and colored LEDs in the eyes for expressive output. The USC Robotics Research Lab currently has 12 NAOs, 10 of which are NAO H25 V 4.0.

The PR2 is an open and robust robot platform designed from the ground up for software developers and researchers. By eliminating the need to first build a hardware system and then re-implement code, the PR2 allows software experts to immediately create new functionality on the robot.

The PR2 robot is fully integrated with ROS, providing the power of all the ROS developer tools and out-of-the-box functionality for everything from full system calibration to manipulation.

We are using the PR2 for a variety of projects, including research in online sensor self-calibration for `power-on-and-go' robotic systems. We are also interested in designing adaptive systems that can facilitate free-form interaction and perform robustly in social situations with children as well as adults, able-bodied as well as those with disabilities. To interact effectively with humans, we aim to make the PR2 possess verbal and nonverbal communication skills, such as speech and body language.

The lab owns 14 Pioneer robots. These include 9 Pioneer 2 DX robots, 2 Pioneer 1S robots, 2 Pioneer 1 AT robots, and 1 Pioneer 2 AT robot. Pioneers are sold by Mobile Robots, Inc.

The Pioneer 2 is differentially steered, has 16 sonar sensors around its 50x50cm body, and contains a Pentium computer running Linux. All our robots are equipped with 802.11 wireless ethernet, a camera/framegrabber and a sound card. Some also carry a SICK laser scanner, bumpers and grippers.

The Pioneer 1 is differentially steered, and has 7 front and side facing sonars. The AT version has fat tires for outdoor use. Our Pioneer 1 ATs carry a lab-built PC104 computer running QNX, with wireless ethernet and Novatel GPS.

Our large collection of pioneers is used to investigate methods for synthesis and analysis of algorithms that enable cooperative behavior. Through both homogenous and hetrogenous teams of the robots a number of interesting and open research questions can be addressed. Certain tasks require more than a single robot, and our large number puts us in a unique position to tackle fundamental problems related to these tasks.

Our robot helicopter research began in 1991 with the formation of the Autonomous Flying Vehicle Project and continues to the present day. During this time our lab has designed, built and conducted research with three robot helicopters. The first robot built at the start of the project was the AFV (Autonomous Flying Vehicle). We transitioned to our second robot, the AVATAR (Autonomous Vehicle Aerial Tracking And Retrieval), in 1994 and to our current robot, the second generation AVATAR (Autonomous Vehicle Aerial Tracking And Reconnassaince), in 1997. The ``R'' in AVATAR changed to reflect a change in robot capabilities.

Since the beginning of the project, a guiding design philosophy has been to create robots with high levels of autonomy. For example, to increase autonomy we require minimal robot dependence upon external resources. Therefore, we locate system power, sensors and computers onboard the robots themselves. Choices made for each are driven by factors such as weight, power consumption and cost. These choices consequently affect the robot control system design. We use a behavior-based control approach for all three robots. As our experience and financial resources have increased over the years, so have the capabilities of our robots. Please follow the links for each individual robot for specifics regarding sensors, computing resources, control system design and other information. (AVATAR Home Page)

Laika and Dogmatix are Sony Aibo dog robots. They each have a total of 20 degrees of freedom, 1 in the mouth, 3 in the neck, 4 in each leg, 1 in each ear and 2 in the tail. Additionally, their paws are passively compliant. Sensors include a color CCD camera, stereo microphones, joint encoders in the legs, touch sensors on the paws, and "petting" sensors on the head and its back. The robot is used in our imitation project that studies the mechanisms underlying the ability to imitate behaviors displayed by others with potentially very different morphologies. Laika and Dogmatix will imitate the movements of a human/humanoid avatar.

Through a collaboration with Evolution Robotics, the lab is supplied with several ER mobile robot platforms. Our primary focus with these platforms has been in Human-Robot Interaction (HRI). These platforms are commercially available; they are equipped with IR's and a camera as the primary sensing modalities. Current work involves building Player-compatible devices drivers and a lab tour guide.

The Segway RMP is a modified version of the Segway Human Transporter designed to provide scientists a mobile base for use in robotics research. These durable platforms have allowed for research in rough terrain while transporting numerous sensing and computing devices. Current work with our two Segway RMP's involve time-optimal outdoor terrain path planning under dynamic constraints, target following, and generation of accurate 3D outdoor maps from laser range data.

Bandit-II is an updated version of Bandit-I. Like its predecessor, Bandit-II's 'body' consists of servo-motors and rapid-prototyped parts, but has a more robust design. Its updated design and fabrication were performed by BlueSky Robotics. It will build upon the techniques used in the successful implementation of Bandit-I to serve as an expressive tool for human-robot interaction. Six Bandit-II robots are currently in use.

Sparky is a Minimatronic Figure (TM) courtesy of Walt Disney Imagineering Research and Development. The 18 d.o.f. robot is tendon-driven using small servos, and can be made to reproduce human-like motions. An immediate application for Sparky will the automated lab tour. By mounting Sparky on a mobile base, it will be used to guide visitors on a tour of the lab. It will stop at various locations of interest to describe, using speech and gestures, the research projects of the lab members. Ultimately, it may also be shown that Sparky's inviting appearance will be useful for other human-robot interaction projects.