No abstract

This paper presents a modular architecture for MAVEN; a mobile robotic avatar. The proposed modular architecture allows for rapid customization to suit changes in application requirements. Aspects of modularity were applied to both hardware and software subsystems in the form of modules for mobility, head and arm gesturing, internet telephony, artificial intelligence (AI) power and communication. Module interconnectivity is achieved using industry standard Ethernet and serial communication standards. This allows modules the flexibility of implementing different operating systems and leveraging on existing development.By grouping hardware and its complementary software into 'self-contained' modules, changes in requirements can be implemented with relative ease; configurations can vary from an intelligent information kiosk to an autonomous mobile Avatar, with arms and head units that permit gesturing. The modular approach allows for the reuse of modules, enhancing ease of reconfiguration and cost savings.

EDGAR is an acronym for "Expression Display & Gesturing Avatar Robot". It is an anthropomorphic telepresence humanoid, which is designed and implemented to allow for social expression. EDGAR is a robotic avatar that was developed by Nanyang Technological University for telepresence applications. It has a total of 28 degrees of freedom and it is able to mimic the remote user's head, torso, arms as well as fingers movements. EDGAR's head is a rearprojection system such that a face texture can be displayed on EDGAR's face. The remote user can use a normal webcam for live-streaming interaction as the face image can be processed to fit the contour of the face. This means that a person's facial features and expressions can be displayed on the robot, making EDGAR suited for portraying any person. EDGAR can also serve as a social robot. A virtual character's face can also be projected on the robot so that EDGAR can be used to perform autonomous social interactions with the people around it.

Lasers are powerful light source. With their thin shafts of bright light and colours, laser beams can provide a dazzling display matching that of outdoor fireworks. With computer assistance, animated laser graphics can generate eye-catching images against a dark sky. Due to technology constraints, laser images are outlines without any interior fill or detail. On a more functional note, lasers assist in the alignment of components, during installation. We propose the use of lasers in the generation of graphics for augmented reality applications. Whilst the unfilled line drawings may be considered as a disadvantage, the ability to project images in a bright outdoor environment is an advantage, particularly in a natural environment. This chapter describes the use of laser outline graphics to augmented reality applications for a proposed 'industrial' application in a sharedenvironment of humans and robots. This implementation demonstrates a novel application and reaffirms the efficacy of laser graphics in providing notification to third party humans in the environment. When a mobile robot is able to indicate its intentions, humans in its vicinity can better accommodate its actions to avoid possible conflicts. A framework for implementing human-robot interface is proposed. Wearable transparent LCD displays offer a high definition graphical interface to the human supervisor to allow for robot control, whilst laser generated notifications allows both the supervisor and other humans in the shared environment to be informed of the robots intentions, without the need of wearable devices. This leads to a more inclusive interaction for all humans in the shared environment.