Is any unmanned spacecraft a robotic probe? You might think so, given the tasks the hardware has to perform to accomplish a given scientific mission, but a more precise definition came out of the European Space Agency's ASTRA 2004 workshop, held in the Netherlands early in November. Says Gianfranco Visentin, head of ESA's Automation and Robotics Section, a space robot is "...a system having mobility and the ability to manipulate objects plus the flexibility to perform any combination of these tasks autonomously or by remote control." And according to this ESA press release, spacecraft robotics should be able to achieve the following: withstand a launch operate under difficult environmental conditions often in remote locations weigh as little as possible as any mass is expensive to launch use little power and have a long operational life operate autonomously be extremely reliable We're gaining a lot of experience with robots through the use of machines like NASA's Mars rovers Spirit...

Apropos of the Cassini material below, IEEE Spectrum Online is running a remarkable story telling how a Swedish engineer discovered a potentially fatal flaw in the communications procedures between Cassini and the Huygens probe that will land on Titan. Corrections to Cassini's trajectory may have saved the mission. Must reading on the subject of spacecraft autonomy and repair.

NASA's Earth Observing One (E0-1) satellite has become a testbed for new technologies. Launched in late 2000, EO-1 is now the venue for a test of artificial intelligence; specifically, an AI software package called Livingstone Version 2 (LV2), which can detect and diagnose simulated failures in systems aboard the satellite. "This is the kind of technology NASA needs to support future exploration of the Earth, moon, Mars and beyond in the 21st Century," said Ghassem Asrar, deputy associate administrator for NASA's Science Mission Directorate. "This software grants us the ability to troubleshoot the robotic systems required to handle increasingly complex tasks of exploration, while they are millions of miles and perhaps light years away from Earth." Precisely. Robotic systems on interstellar missions -- and 'interstellar' also means relatively 'nearby' destinations like the Kuiper Belt -- will of necessity be autonomous and self-correcting. Be sure to read Greg Bear's Queen of Angels...

Carnegie Mellon University is about to host the 25th anniversary celebration of its Robotics Institute. A four-day symposium will begin October 11 to discuss the grand challenges of robotics and the commercialization of robotics research. October 13's lineup is particularly stellar: Vernor Vinge, professor emeritus, University of California, Davis, known for his science fiction, including True Names and Marooned in Realtime, speaking on "Robotics and the Technical Singularity." Robin Murphy, professor, University of South Florida, an expert in search-and-rescue robotics, speaking on "Up from the Rubble." Bob Full, Chancellor's Professor, University of California, Berkeley, speaking on "Bipedal Bugs, Galloping Ghosts and Gripping Geckos: BioInspiration in the Age of Integration." Mitsuo Kawato, director of the Computational Neuroscience Laboratories of the Advanced Telecommunications Research International, whose approach is that "we construct a brain in order to understand the...

An interstellar robotic probe will need robotic systems that are completely autonomous. Think about it: the Jet Propulsion Laboratory was able to solve the rover Spirit's problems on Mars by sending it a series of commands and a software patch. Spirit's computer was crashing whenever it tried to access its data storage, forcing a series of reboots that were running down the rover's batteries. By deleting a backlog of files that were clogging the system's flash memory, JPL got the rover back on track. But Mars is virtually next-door compared to Alpha Centauri. Even the Galileo probe, which experienced a tape recorder malfunction and a reprogramming forced by the failure of its high-gain antenna, could be fixed from Earth because the radio delay was only eighty minutes round-trip. Imagine what will happen with an 8.6 year round-trip delay and you can see why robotic repair systems will have to function on their own. Of course, not all progress on robot autonomy is related to space...

Build a model to test out systems on an aircraft and you've created what engineers call an 'iron bird.' Now, as this press release from NASA explains, the agency is developing a 'virtual iron bird' (VIB) that would computerize the process, providing three-dimensional images of of individual systems and parts that engineers or astronauts can use to debug problems or refine improvements. In the work, being conducted at NASA's Ames Research Center in California, the VIB is envisioned as a complete computerized model of the spacecraft that would be carried in its computers, including engineering drawings, specifications, part numbers and all relevant details. Another use of VIB will be to simulate flight situations for vehicles still in the planning stages. Centauri Dreams' take: we'll need onboard systems that include evolutionary algorithms aboard interstellar probes, allowing them to become self-healing and react to failures of components in ways that are more flexible than current...

NASA's Spirit and Opportunity rovers have given us remarkable views of Mars, but the next generation of such vehicles will need greater onboard intelligence. And as we move ever farther away from the Earth, our systems will eventually need to become capable of a great degree of autonomy. Imagine an Alpha Centauri probe 4.3 light years away as it experiences a malfunction, or discovers a new target in Centauri space that needs investigation. There will be no way to wait 8.6 years for a round-trip signal to Earth, so autonomy and AI are crucial for robotic exploration. A recent story in SpaceDaily talks about NASA's work at its Ames Research Center that focuses on improving what our rovers can currently do. An interesting quote from NASA's Kanna Rajan: "Part of the problem is we are not closing the loop on board the rover. Signals from the rover have to go to Earth for a human with his/her cognitive capacity to deliberate on the information in the signal and make a decision. Based on...