Bristol Wearable Computing

[ Home | Plans | Meetings | Members | Search | Papers | Links | CyberJacket | LocoMedia]


Project Meeting: 18th August 1998

At: 9:30am, MVB 3.35

Present: David M, Henk, Phil S, Dave R, Marianne, Jim

Annual Review

  • Initiated 3 fourth year projects in 97/98 (SitComp Sever, Virtual sensors, Motion sensing)
  • Sought wearable RA
    • placed 3 ads in national press,
    • advertised in newsgroups and Email lists
    • interviewed 5 candidates, made two offers.
  • Designed and planned LocoMedia Summer Project covering themes:
    • environmentally aware
    • dynamically connected
    • unconsciously wearable
    • personal policies
  • Thought up and started building Foot-Bridge short range comms link.
  • Increased profile within the community
    • presented at IEE Symposium on Personal Technologies (Savoy Place)
    • presented FootBridge at ISWC '98 (Pittsburgh)
    • attended 'HCI for Mobile devices' symposium (Glasgow)
    • personal contacts at BT, Philips, UCL, Uni Kent, Uni Washington, Uni Oregon
    • contact with other HP groups (HPL Mark Smith, XyberNaut, State Farm).
  • Conducted LocoMedia Project
    • 5 students working in new Info App Lab at MVB using PC104 hardware
    • incorporated HP speech technology, FootBridge and SitComp server.

What we learnt

  • Its hard to operate without a full time researcher driving the programme at the Uni. jbr, hm
  • We need better hardware support. Find more of Phil's time? Involve the EE department more. Outsource more. jbr
  • LocoMedia was slightly overambitious. Subtle, critical design issues permeated the system. It was hard to keep control of these issues while still allowing students freedom to do lower-level design as they liked. Use RA for more platform work. jbr
  • Its hard to attract RA applicants with broad systems skills - there's too much competition from industry. Its hard to attract applicants to a new field. We must revise expectations downwards or look for funds for a chair?! jbr, dm
  • We need to pay more attention to robustness issues. dm
  • Its unrealistic to expect to be able to direct students accurately. jbr

Discussion points

  • It is essential to keep Mike (one of the five Summer Students - retained one day a week this term) close to the project for at least the remainder of the year. Mike is doing really excellent work on the platform and he is crucial to the success of the wearable programme.
  • We might consider getting Phil to collaborate with Bristol EE guys in an effort to support his near-field work and ideas.
  • We need to eliminate HW issues from platform debugging. Develop  HW and SW components to check the functioning of the platform.
  • We need a systematic approach to deal with audio robustness:
    • Track s/n ratio on sound card.
    • Develop mic set up programme
    • Measure how people use speech. Use this to tailor vocabulary and to redesign dialog.
    • Could solve this by Marianne directing RA?
  • We could use EE skills to develop fast wireless comms and then emphasise client/server themes in wearables.
  • We should be able to solve current platform power supply problems. Wires are too thin. Could use PC104 power management boards, but then we would need to take power off the PC104 to power other devices (GPS). jbr thinks the basic PC104 set up should be workable, given the experience of MIT with their Lizzy system. Note that they use the same sound card as us, though their whole system is better packaged in a single unit.

Plans

  • Set RA on programme to improve robustness of platform. Nov 98
  • LocoMedia demo. Xmas 98
  • Involve EE dept in research. Jan 99
  • Consider changing HW platform. Mar 99
  • Stable, documented development platform supported by an RA. (realistic?) Summer 99.
  • Collaborative and social applications built and tested by students. Oct 99.
  • Visits to other wearable groups (BT, Oregon?) Oct 99.
  • LocoMedia presented to the broader wearable community. Oct 99.
  • More partners added to BWCI, EU funding. 2000

Actions

  • Contact Cliff to get him on board. dm
  • Pay Mike! hm
  • Arrange meeting to start him on programme of platform robustness development. jbr/hm
  • Contact EE department to get them involved in programme. dm
  • Set up meeting to discuss research agenda with EE folks. Get Phil in. jbr
  • Think up necessary elements in a 'Cyber Jacket robustness checker' system. hm
  • Phil finish FootBridge. pin
  • Get latest speech recognition system going on CJ.

[Next meeting | Previous meeting | Meetings | Back to top]


ęCopyright Hewlett-Packard 1997, 1998.
The material displayed is provided 'as is' and is subject to use restrictions. For problems or queries about this web contact web master. Last updated: 14 Jan 2000.