Search results

Service robots need to be programmable by their users who are in general unskilled in the art of robot programming. We have explored the use of spoken language for programming robots.

Two applications domains were studied: that of route instructions and that of game instructions. The latter is work in progress. In both cases work started by recording verbal instructions representative of how human users would naturally address their robot.

The analysis of these instructions reveals references to high‐level functions natural to humans but challenging for designers of robots. The instruction structure reflects assumptions about the cognitive abilities of the listener and it is likely that some human capabilities for rational thinking will be required in service robots.

Some of the high‐level functions called for by natural communication stretch current capabilities and there is a clear case for more effort being devoted in some areas. Instruction analysis provides pointers to such research topics.

It is proposed that service robot design should start with investigating the way end‐users will communicate with the robot. This is encapsulated in the “corpus‐based” approach to robot design illustrated in this paper. This results in more functional service robots.

The paper stresses the importance of considering human‐robot communication early in the robot design process.

This paper reports a few results of an ongoing research project that aims to explore ways to command an industrial robot using the human voice. This feature can be interesting with several industrial, laboratory and clean‐room applications, where a close cooperation between robots and humans is desirable.

A demonstration is presented using two industrial robots and a personal computer (PC) equipped with a sound board and a headset microphone. The demonstration was coded using the Microsoft Visual Basic and C#.NET 2003 and associated with two simple robot applications: one capable of picking‐and‐placing objects and going to predefined positions, and the other capable of performing a simple linear weld on a work‐piece. The speech recognition grammar is specified using the grammar builder from the Microsoft Speech SDK 5.1. The paper also introduces the concepts of text‐to‐speech translation and voice recognition, and shows how these features can be used with applications built using the Microsoft.NET framework.

Two simple examples designed to operate with a state‐of‐the‐art industrial robot manipulator are then built to demonstrate the applicability to laboratory and industrial applications. The paper is very detailed in showing implementation aspects enabling the reader to explore immediately from the presented concepts and tools. Namely, the connection between the PC and the robot is explained in detail since it was built using a RPC socket mechanism completely developed from the scratch.

Finally, the paper discusses application to industrial cases where close cooperation between humans and robots is necessary.

The presented code and examples, along with the fairly interesting and reliable results, indicate clearly that the technology is suitable for industrial utilization.

Aims to demonstrate how robot technology, machine vision, voice recognition and artificial intelligence can be applied to creating an automated surgeon's assistant that is functional and cost‐effective.

Presents the development process that led to the construction of the Penelope Surgical Instrument Server (SIS), outlines the mechanical design of the robot, describes the control strategy and reports on the first real live implementation in an operating room. Machine vision, voice recognition and artificial intelligence are combined to create a robot assistant that is able to anticipate a specific surgeon's needs for a specific surgical procedure.

Finds that a robot can manage an array of surgical instruments and present the right one to the surgeon at the right time.

A robot for assisting a surgeon can relieve support staff in hospital operating rooms of repetitive tasks and thereby improve patient care.

Introduces the concept of a robot for assisting surgeons in operating rooms and freeing up scarce financial and human resources to provide for better nursing coverage in patient care areas.

Aims to summarise the programme of UV Europe, a conference on unmanned vehicles that took place in Paris in June 2005.

Selects key conference presentations and summarises their contents, indicating the main themes.

Finds very strong international activity in the field of unmanned aerial vehicles (UAVs). Most progress has been made in the military sector so far. Concern about air space safety is holding back the industry.

Presents the topics relating to UAVs that concern the market today, and summarises the main areas of research and development.

This paper aims to design a new full‐body tactile sensor which is essential for the application of personal service robot similar to human skin.

The largest difficulty for designing a full‐body tactile sensor is the huge number of output connections. The sensor introduced in this paper is a special multi‐layer structure, which could minimize the output connections while sensing both the position and force information. Since it is made of conductive and non‐conductive textiles, the sensor could be used to cover the curved surface of robot body.

With better structure design, output connectors and signal measurement times could be dramatically reduced.

Sensor area and performance are limited by the sensitivity of the measurement circuits.

Introduces an innovate design of full‐body tactile sensor.

Examines a recently launched integration of smart cameras into industrial robots to make them responsive to a changing environment.

Reviews the capabilities of the vision‐enabled robot, citing installations in Sweden and the UK, then describes the robot and vision programming procedure.

Vision integration opens up a range of new possibilities such as simultaneous product handling and inspection, as well as providing real‐time robot guidance. Standardisation plays an extremely valuable role in building integrated systems from disparate technological elements. Here ActiveX web standards, ethernet connectivity, a standard interchangeable family of cameras and a common controller for a whole range of robots are the keys to the synthesis of a powerful new combination of robot and machine vision.

Draws to the attention of industrial engineers the availability of a family of robots with integrated machine vision.

Aims to collate and analyse data relating to the sales and installations of industrial robots in the UK for the preceding year and, by comparison with data from previous years, identify market trends.

Data has been collected by the annual survey of UK robot sales and installations conducted by the British Automation and Robotics Association.

The key statistics for 2004 are as follows: a total of 785 units installed; total installations to date (from 1978) are 21,744 units adjusted to 14,176 to allow for International Federation of Robotics 12‐year retirement rule; the automotive sector remains dominant but dropped 79 per cent from 2003 to 458 units; the non‐automotive sector grew 11 per cent from 2003 with 327 units installed; significant growth in three sectors are: plastics and rubber 46 per cent; construction 85 per cent; and pharmaceuticals 95 per cent. Arc and spot welding installations reduced by 45 per cent, however, spot welding remains dominant application; growth of 80 per cent in “Dispensing” applications; six‐axis robots account for majority of the market; significant growth in seven‐axis robots; majority of installations fall between £10 and 30 k cost classification highlighting a price drop from 2003; robots originating from EU decreased by 145 per cent; and robots originating from Japan increased by 8 per cent.

The study identifies current trends in the UK industrial robot market in terms of factors such as industrial sector, application, robot type/structure and cost. By relating the data to the economic and industrial conditions prevalent in the UK this will provide a useful insight for robot application and development for both end‐users and vendors.

To report on the design and development of a robotic automobile aluminium frame welding system.

The customer contracted with a team of suppliers, each with unique skills and products necessary for welding thin aluminium framing. The team employed simulation to help in the design process to avoid problems and to speed the time to complete.

Careful planning, detailed simulation and attention to details helped insure that working production systems was in service to meet a tight deadline.

Companies looking to develop welding systems for structures from aluminium or other materials can learn from the benefits of a team approach to complex application development. As the auto industry moves to more use of lightweight materials such as alumnus, successful automation of its welding will be ever widely appreciated.

Any user planning an automated aluminium frame welding system can learn from the success of Ford Motor and its welding system design partners. Careful planning and teamwork can help in meeting production and timetable goals.