Baby X – Spatial cognitive journeys into a child’s mind

Wednesday, 14 January 2015 12:00 AM

The spatial industry is rapidly blurring our physical and digital realities to better enable our comprehension of the world around us and our deeper understanding of one another. There is a soft continuum between the ‘Digital X’ domains of medicine and geosciences. All share the common need to correlate position in both time and space, and to utilise visualisation to render metaphors for our mind to comprehend vast complex data. Algorithms, data wrangling, analytics, models and even policy and standards cross this divide between between a deep implicit understanding of ourselves and the planet surrounding us. Nature similarly does not recognise such boundaries – the man built environment and the natural world are intricately intertwined systems of a whole. Spatial technology is now enabling us to break down knowledge siloes of the past and work meaningfully with the bigger picture. This break through is critical for knowledge amplification and solving some of the greatest challenges facing humanity.
A most striking project that demonstrates this potential is BabyX of Auckland's Bioengineering Institute Laboratory for Animate Technologies, for the creation of a virtual animated baby that learns and reacts like a human baby. It uses the computer's cameras for "seeing" and microphones to "listen" as the inputs. The computer uses Artificial intelligence algorithms for BabyX's "learning" and interpretation of the inputs (voice and image) to understand the situation. The result is a virtual toddler that can learn to read, recognize objects and "understand." The output is the baby's face that can "speak" and express its mood by facial expressions (such as smile and show embarrassment).

Baby X’s first words

Advancing horizons in the spatial industry

SIBA's 'Spatial Data Ecology' diagram illustrates a data value chain for the spatial industry.  Commencing with raw data value is built by structuring this into information models; utilisating big data veracity and metadata such as provenance to build greater trust in the data as evidence; simulating the model against natural world phenomena to build evidence into deeper knowledge; applying artificial intelligence to this knowledge to fill in the pattern gaps; and finally the visualised expression of this rich information through metaphors to spark neurons in the human mind to make a decision or create an idea to drive ongoing iterations of this ecology.  Activities of the spatial industry are mapped to each step in this journey. These activities map to companies and illustrate their role in the overall value building. Companies can then be mapped to industry and competive/cluster models to develop a deeper understanding of the spatial industry and its relevance as an enabling infrrastructure to emerging business models. 

THe cycle below shows the heart of this Spatial Data Ecology. The reality today is that this value chain is short circuited - data is published directly to visualisations (metaphors), or at best it may be modelled (eg. with BIM)  and structured to some superficial extent. 

Through project MIMOSA, SIBA is working with the spatial industry to steward the shift from map-making to big data and seeking to building the value through collobration and the context of a Digital Earth like open innovative forum to fulfill the complete ecology.  Once this context  is achieved the spatial industry will be contributing to a knowledge amplifier.  In February 2015, SIBA and QUT will be hosting a Geobrowser workshop and inviting the world's experts in this field to Brisbane to participate. There is excitement of the potential of what the Federal Government's National Map initiative. This initaitve has adopted Cesium as an open source platform to publish the the Government's open spatial data to.  The open context as well as open content is the foundation for building a Digital Earth knowledge amplifier.  The Queensland Globe has been a pioneer is demonstrating the potential of brining together disparate spatial data from across Governent and delivering it to public and professional users through a common browser.  The earlier Queensland Globe adopted the more proprietary Google Earth as its platform as options were limited at the time.   The Geobrower workshop will help identify a strategic road map to advance the Digital Earth context.

The Auckland Bioengineering Institute is a leading participant in a world wide initiative known as the Physiome Project.  This initiative is akin to Digital Earth in some respects but at the scale of a human from the protein level to the full body (physiome).  The physiome project models have application in drug discovery and drug testing as well as to the applications illustrated by Baby X.  Research outcomes are published directly into a common model for curation and and contesting by scientists.  A good description of the Physiome project is in this Nature paper  by Professor Peter Hunter and Thomas Borg.  The Digital Earth vision is explained in this paper by Mike Goodchild et al that was published in the Proceedings of the National Academy of Sciences (PNAS).

In the spatial industry applications such as  mining industry and robotics have  grappled with concepts such as Latent Truth. Mineral exploration has extensively applied methodologies such as Krigging and Evidence Belief Functions to help in forensic identification of spatial resources . Artificial Intelligence methods such Bayesian Nets and other techniques applied to the forensic discovery of truth and analytics of Big Spatial Data models will be a major growth area of innovation in the spatial industry over the coming decade as the Digital Earth contextual models become more established and supported.

The Baby X projects is an excellent example to demonstrate this fusion of artificial inteligence into spatial technologies. Dr Mark Sagar heads artificial intelligence research in this area at the University of Auckland's Bioengineering Institute. Mark is known for being a two-time Oscar winner for his work with Weta Digital.

Dr Mark Sagar was one of the expert panelist in the Royal Society of New Zealand's forum on  ' Big Data: Changing Place'.  Mark Sagar, SIBA's Richard Simpson, and one of NZ's leading Research Geologist Dr Cornel de Ronde were interviewed  in a panel discussion by Kim Hill, an international award winning journalist in 2013 for New Zealand National Radio.They jointly tackled one of  the exhibition's themes 'The sentient planet – technology as a super sense'. The human body has a limited number of senses. Technology enables us to perceive and act beyond these limits and blur the boundaries between the natural and virtual worlds. We can now see what was previously invisible. What does this mean for our future? 

Richard Simpson was the curator of the "Big Data: Changing Place" exhibition for the reopening of the New Zealand National Library.  At one point in the planning of the exhibition there was an  idea  considered to include an early demonstration of Baby X to demonstate the parallels in the modelling the body and the planet, however alternative options were substituted as it was too early in the project development to meet the deadlines.  Simpson was formerly head of innovation at the Auckland Bioengineering Institute, and the exhibition crossed  experiences working with cutting edge bio-technology and geo-technology. One intriguing ABI work included was the world's first  - a Turing Machine made entirely from artificial muscles. This was created by Dr Ben O'Brien of ABI, and possibly hints at what may be yet to come in the evolution of Baby X.

Public Launch of Baby X at TED X in Auckland

The Auckland Bioengineering Institute
The Auckland Bioengineering Institute (ABI) is a cross-faculty research centre with the world wide academic community.  ABI’s research deals with the application of mathematical and engineering sciences to biology and human physiology. The aim of this research is to improve understanding of physiological processes and the diagnosis and treatment of injury or disease. Many of the staff have joint appointments in other Faculties, especially with the Department of Engineering Science in the Faculty of Engineering and the Department of Physiology in the Faculty of Medical and Health Sciences.

Spatial Industry Innovation model
In 2007 “The Innovation Pathway for Bioengineering” grant was awarded to fund a scheme called the Growth and Innovation Pilot Initiative (GIPI) scheme. The high level objective is to mesh business needs with research capabilities through human capability building and knowledge sharing. While a major part of the pilot relates to engaging with industry on the institute’s core capabilities (initial core capability selected was musculoskeletal modelling), there is also strong  support for entrepreneurial activities through the Knowledge Transfer Fund.

This initiative linked the world leading research capabilities of the ABI with the emerging biotechnology industry in New Zealand. It successfully sought to ignite a culture of entrepreneurship developing within the Institute and the wider University, and gain leverage from support available through Auckland UniServices Limited, the commercial arm of the University of Auckland. The overriding objective of the initiative is to mesh business needs with research capabilities through human capability building and knowledge sharing.

Inspired by phenomenal success ABI's GIPI , SIBA has established the Spatial Innovation Foundation (SIF). The beauty of SIBA’s SIF approach is it agile approach for stimulating new innoation and businees by engaging cross-disciplined discussions with business, research sectors, policy makers, and data providers. SIF addresses the place where the market is broken and supports the creation innovative platforms for advancing the knowledge economy. 

ABI's Baby X is a good example of a very complex concept that would not have got off the ground without stewarding through a feasibility prototype stage and bringing together many different skills.  It is also an example to demonstrate how a small nation with a tiny internal market such as New Zealand can strategically build local IP and retain and attract  international expertise to build a basis for a domestic high tech industry.  SIF has the pedigree of the experience in setting up GIPI and is fuelled by a need to address market survival needs as determined by business rather than academia or Government. SIF is driven by the private sector, rather than the research sector, as it addresses the real business needs of the emerging spatial industry and the critical importance innovation has for business in surviving  paradigm shifts.
Equivalents to Baby X in the spatial industry are projects SIBA is enganing in is Project MIMOSA and VANZI.  These are tightly coupled and both have high dependency upon challenging policy, technology innovation, and demand for digital content. Like Baby X they potentially offer a new platform for knowledge amplification and open the doors to new and better ways for achieving digital cities and growing export opportunities in the spatial industry



no comments

Login to add comment

Recent Articles