Streamlining the warehouse: Xcon and Daimler's Xpick picking system
Wearable computing systems are at long last becoming a practical option, after years when they have typically been ungainly and difficult to integrate in industrial environments. The shift is due to a convergence of better design, pervasive wireless, ubiquitous smartphones and other mobile developments such as tablet PCs.
The technology for wearable computing has been in development for 20 years. It has taken the form of head-mounted displays (HMD) and sensor arrays for 3D design of cars and aircraft, virtual reality, surgery and military applications, and smart clothing for animation, gait analysis and medical diagnostics. Recently though, at the Consumer Electronics Show (CES) in Las Vegas, the technology was creating a buzz, with predictions that it would affect a spectrum of industrial applications rather than just gamers, geeks and fitness freaks.
The market is being driven by leading players, such as Google with Google Glass and Apple, and enterprise computing majors such as SAP are taking wearable computing seriously. Developments in key technologies such as augmented reality offer potential benefits in warehouse management, production line optimisation, maintenance and training. Wearable sensors could also revolutionise medical home diagnostics and care of the aged.
One initiative looking at industrial and healthcare applications is the European Commission’s wearIT@work programme, which has 42 partners and funding of €14 million to investigate four industrial pilot applications including production, maintenance, healthcare and emergency services. Pilots have typically run for 18 months since 2004 in industrial environments, including two Skoda car production plants in the Czech Republic.
The pilots showed that wearable computers prevented mistakes, helped people work faster and more efficiently, and shortened the training process for assembly workers involved in tasks such as front headlight assembly.
Tests were carried out using a Xybernaut V computer, a Microoptical SV6 HMD and a conventional headset, with a VNC client application, to allow remote control and interaction with a wearable application for 14 assembly procedure steps. The tests highlighted the need for the wearable devices to be unobtrusive, for good multimodal interaction, for a collaboration mechanism, and for task tracking via mobile sensors to ensure tasks were performed correctly, providing help and suggestions via the headset.
Complex maintenance tasks were also studied at Airbus and EADS facilities in partnership with Giunti Labs in Italy, to demonstrate how systems could increase the mobility of workers in geographically dispersed production plants and improve availability of task information, knowledge sharing and training, with multimodal interaction through voice and gestures.
Maintenance operators were equipped with a modular maintenance vest configured with cutting-edge hardware components, input and output devices for hands-free working, voice and gesture-based interaction, with advanced knowledge management functionality. Team members with remote support claimed that work quality and cooperation improved with new audio-visual tools and on-the-job decision support.
The wearable system provided better access to documentation and company know-how, with significant time reduction for complex tasks, improved quality and reduced errors. The Wearable Maintenance Integrated Solution (WMIS) was open, flexible and based on a modular wearable platform.
On the healthcare front, wearable technology can improve the availability of information for doctors and nurses on ward rounds. Studies in this field were carried out by Austrian TechCenter, Gespag Oö, in concert with Professor Michael Lawo at the TZI Centre for Computing Technologies at the University of Bremen. Wearable computers were equipped with input devices for gesture recognition, RFID scanners, and proximity sensors for communication with fixed infrastructure, such as bedside displays, and devices used by medical colleagues, such as tablet PCs.
Harold Reiter, principal scientist at Philips Healthcare in Eindhoven in the Netherlands, has been looking at wearable systems for home monitoring and diagnosis since 2000. He sees “significant opportunities for robust, reliable wearable solutions for monitoring the health of patients suffering chronic conditions”.
The company was involved in the ¤35 million MyHeart and HeartCycle programmes, in partnership with 33 industrial, research and medical organisations, to develop preventative and management technology to monitor vital body signs for direct feedback to users or physicians.
Philips has been developing a body vest with integrated textile electrodes and control electronics to measure the patient’s ECG. The devices transmit measurement data via wireless links to a PDA on which heart failure management software runs. Smart textiles can integrate monitoring sensors and electronics unobtrusively, allowing home monitoring of ECG and other biomedical measurements that would normally be measured only in a hospital environment, says Reiter. “The main challenge is to facilitate self-measurement with a robust and reliable system that will integrate in daily life. False alarms must be avoided, with fast access to reduce time to treatment.”
Smart textiles typically involve a lot of hand-made work, but for the MyHeart programme Philips worked with an industrial knitting machine company to produce smart textiles for mass production. The company is collaborating with Finnish firm Clothing Plus, which has production facilities in China, to produce a standard textile, incorporating complex electronics, that is machine-washable 50 times at 40°C. Connected research is also underway in Philips laboratories in Cambridge, at the University of Hull and Imperial College.
Meanwhile, many auto manufacturers, including BMW, Audi, Daimler and Volkswagen, are pursuing opportunities with wearable computing, says Lawo, who oversaw the wearIT@work programme. “Most of the work is secret, because the impact on production efficiency and the increase in quality is tremendous,” he adds.
The US took up the wearable computing opportunity before Europe, says Lawo. “The breakthrough came with the wireless head-mounted display. Unfortunately, Europe missed a great business opportunity – despite initiatives by Trivisio and TeXXmo – as US developers took up the challenge, with systems such as Google Glass on the consumer side, and Motorola’s HC1, Vuzix’s M100 and others on the industrial side.”
The main challenge is how information via the wearable system can be integrated into the overall IT infrastructure, says Lawo. “It’s not simply a matter of transferring the desktop interface onto an HMD and using speech input instead of typing. You need a carefully integrated system with a specific text or graphical user interface.”
That challenge explains why SAP is running with the ball. As an enterprise provider, it has lots of pre-processing and post-processing partners and is encouraging them to develop wearable applications. An interesting YouTube video with SAP and Vuzix demonstrates a warehouse picker getting directions on his M100 smart glasses, using augmented reality.
Vuzix recently introduced the M100 monocular display, which uses the Android operating system, and provides an integrated camera, audio system and processor in a neat, hands-free format. The system operates with most Android apps and is Bluetooth- and WiFi-supported, to run with or without cloud connection. Industry has been the focus for Vuzix chief executive officer Paul Travers since research into virtual reality in the early 1990s, after which his company became a supplier of thermal night vision systems for Raytheon and DRS.
The M100 can be used with an adjustable over-the-head band or clipped-on safety glasses to access, record and review data. Travers says formatting information is important for a crisp display, with the opportunity for optical see-through to the object being assembled or picked. Vuzix, in partnership with Nokia, is also introducing a super-thin Waveguide lens that can project full-colour, augmented-reality images over surroundings, using V2000AR glasses.
Another challenge for wearable computing is to provide access to information that can be consumed easily, says Lawo. “We need to improve hypertext reading dramatically – much as Wikipaedia was such an advance on the paper version of the Encyclopaedia Britannica – using hyperlinks, as wearables depend on multi-media and audio information.”
He suggests the next step is ‘hyper-media’. “The big issue is not technology but how to integrate wearable computing into the existing processes and software environment. If you can show decent return on investment, I’m convinced wearable computing is the next step after the mobile revolution.”
Dr Hendrick Witt, chief executive of Xcon Partners, runs a wearable computing competence centre in Bremen, Germany. He believes “there are many application domains with business potential for wearable computing, including hands-free information-driven maintenance and service, visual support for complex production procedures and logistics, healthcare, security and support for disabled people”.
Witt’s company, with Daimler and other partners, has developed a hands-free, multi-order, picking system called Xpick. It can be used with an HMD, tablet PC on a picking cart or on a wrist-worn computer. A special user interface can be used with different picking error reduction modules for intelligent pick weight matching, cart location, integrated barcode scanning and voice-based operation.
“Graphical visualisation on HMD shows lower error rate and faster work order times compared with paper- and voice-based solutions,” says Witt. The HMD uses a wireless Lan with various picking scenarios and a graphical user interface. Daimler is testing prototypes for warehouse management and logistics, and the system was launched in February at the LogiMAT logistics show in Stuttgart.
The wearability of the technology has come a long way. In the early days of VR, there were concerns about extensive use of of HMDs and repetitive strain injury. Today’s systems are lighter, neater and more flexible. Workers take a little time to adapt to head-mounted systems, but are generally comfortable after a short time, says Witt.
Driver dynamics: Xsens system with Delmia software
Another company, Xsens in the Netherlands, specialises in 3D motion tracking systems for humans and machines for industry, entertainment and movement science. The Xsens MVN is a camera-less system, which has 17 inertial sensors for motion capture and is used for digital posture and gait measurement by industry, film and animation. Rob Loring, sales manager for measurement science, says: “Wearable computing is at a turning point.”
Xsens’ systems are wireless motion trackers equipped with MEMS (micro electronic mechanical systems) sensors. Inside each Xsens tracker are three MEMS accelerometers, three MEMS gyroscopes and a magnetometer. The motion trackers can calculate the 3D orientation of each tracker and where it is heading. “The information is run into a computer, where a biomechanical model defines the orientation of the model or animated character,” says Loring.
Full-body versions of Xsens’ system are being used by large automotive production companies, including Fiat and Jaguar Land Rover, for ergonomic analysis of new production lines and for assessing human interaction with new car designs, in combination with Catia or Delmia software and Siemens’ Jack or Process Simulate Human systems. The Xsens system is also used by aircraft manufacturers, including Embraer in Brazil, to design new production lines.
Though Xsens offers a tight-fitting, full-body, Lycra suit for animation and gaming, its latest Mark 4 sensors simply strap on to the body for industrial applications, with full calibration in a few minutes. The system can also be hidden under a jacket.
Indeed, industry looks set to become far more comfortable in general with wearable computing. As with the mobile computing explosion, awareness is being led by the consumer arena and by innovations such as Google Glass. But industry adoption will be driven by economics, efficiency and the advantages for order picking, maintenance, inspection, safety and leaner operations.