By Jonathan Amos
Science reporter, BBC News
Europe's proposed satellite-navigation system, Galileo, faces big delays and cost overruns unless major obstacles to its development are removed - and fast.
So far, only test spacecraft for Galileo have been launched
A large order for spacecraft must be placed in the coming months if the project is to keep to a 2011-12 target for full operational deployment.
But negotiations to set up the private framework that will implement and run the system have now been suspended.
It has fallen to the German government to try to break the impasse.
The Germans currently hold the European Presidency, and its Transport Minister Wolfgang Tiefensee is set to chair a number of critical meetings.
One of these is the gathering of his EU counterparts in Brussels next week.
On the agenda will be a report about the glacial progress that has been made towards establishing the private company which will buy most of Galileo's infrastructure and sell its services.
This is supposed to be a "merged consortium" comprising some of the biggest names in the European aerospace industry: EADS, Thales, Inmarsat, Alcatel-Lucent, Finmeccanica, AENA, Hispasat, and an eighth partner that includes Deutsche Telekom and the German Aerospace Centre.
Since June 2005, public officials have been negotiating with this group to put in place the details of a 20-year concession.
GALILEO UNDER CONSTRUCTION
A European Commission and European Space Agency project
30 satellites to be launched in batches by end of 2011-12
Will work alongside US GPS and Russian Glonass systems
Promises real-time positioning down to less than a metre
Guaranteed under all but most extreme circumstances
Suitable for safety-critical roles where lives depend on service
But these discussions are now on hold because the consortium itself cannot agree a common commercial position.
And in addition to these differences, there are also complaints of political meddling, with EU member states still pushing for their interests to be made paramount. Arguments continue over where control centres should be sited and where industrial contracts should be placed.
"There is a minority of shareholders inside the consortium who are strongly advocating what they feel are their national and commercial interests, to the detriment, quite frankly, of the overall project," explained Ruy Pinto, director of satellite operations and navigation at Inmarsat.
"I can say that comfortably because Inmarsat has constantly advocated that we move forward and treat this as a European project. We are advocating a compromise that makes business sense and does not delay the project further," he told BBC News.
The intention behind Galileo was that taxpayers in the EU would inject more than one billion euros (£0.7bn) into the early development of the project.
The deployment phase - the launch of the satellites and the construction of ground stations - was expected to cost at least two billion more, with two-thirds of the investment being borne by the private sector.
The latter was also expected to pick up all the running costs in the long term.
The first demonstrator spacecraft, Giove-A, is already in orbit. A second, Giove-B, which has had some technical problems, should be in orbit by the year's end.
And a third test platform, Giove-A2, has also just been ordered from manufacturer Surrey Satellite Technology Limited in the UK.
These demonstrators are being lofted to secure the radio frequencies allocated by the international community to Galileo and to allow in-orbit testing of key technologies, such as the atomic clocks that are essential to the operation of a sat-nav system.
The contract for the first four satellites proper in the final constellation was awarded at the end of 2004, but the time has now come to order the remaining 26.
If some of this order is not placed in the next few months, a major gap will develop in the deployment phase, leading to a big delay in the completion and operation of Galileo.
It should be the private concession making the spacecraft order - but it does not formally exist yet.
Pedro Pedreira runs the GSA (GNSS Supervisory Authority), the agency tasked by the EU with trying to seal the public-private partnership (PPP) and then act as watchdog to the newly formed Galileo operations company.
He told BBC News that to date there had simply not been enough communication and flexibility on the part of many involved in the discussions. And he warned that unless matters changed soon, the whole PPP approach might have to be reviewed.
"We have given an opportunity to this model of partnership and if it doesn't work - and we will continue to try to make it work and we believe we can make it work - then we have to move on because it is too important for Europe to lose this chance," he said.
"We are in danger of delaying the full operational capability (FOC) of the system because we are discussing details that will not impact on FOC but will only be relevant many years afterwards.
"We have to break down this agreement into phases and start with what is very urgent; and urgent is everything that has an impact on the FOC of the system."
Timeline slippage has been a constant feature of Galileo. When the project was given its final go-ahead by European ministers in 2004, the date of 2008 was still being touted for FOC. This was later moved to 2010; it has now gone out to 2011-12.
When the system does finally get into orbit, it is expected to drive a multi-billion euro industry in which receivers find their way into many more markets - from consumer mobile devices, such as phones, to safety-critical applications, such as guided trains and buses.
Its next-generation technologies promise performance improvements that should see sat-nav receivers get a fix in even the deepest "urban canyons".
Satellite navigation systems determine a position by measuring the distances to a number of known locations - the Galileo satellites
The distance to one satellite defines a sphere of possible solutions; the distances to four satellites defines a single, common area
The accuracy of the distance measurements determines how small the common area is and thus the accuracy of the final location
In practice, a receiver captures atomic-clock time signals sent from the satellites and converts them into the respective distances
The whole system is monitored from the ground to ensure satellite clocks do not drift and give out misleading timings