Recent discoveries have fueled our interest in the possibility of life elsewhere: the presence of ice on our moon's poles by the Lunar Prospector; the Galileo's evidence of oceans on Europa (and now Callisto too!); and the Hubble Space Telescope's views of collapsing protostars in the Orion nebula. These tools have allowed us to expand our understanding of the universe at a pace and scope unmatched by any previous generation. These tools have also allowed us to act upon our dream of actually locating life far beyond our own birthplace.

As we expand outward in this search, how will we get to all the places we need to go? And who will "we" be? It would be tempting to extrapolate our recent successes in designing ever more capable and intelligent robots to think that we, flesh and blood humans, won't have a place in the search for life in the universe. In this vision of the future, like armchair quarterbacks, we'd sit staring at our holographic monitors of the future, cheering on our robots, with us as mere couch potatoes looking on at their successes. What an incredible cosmic irony, to think that just at the point where Earth's own life is about to venture forth in the most fundamental search of all time, we'd chose to stay behind.

Robots are great at doing pre-programmed tasks, and we've now built a new generation of robots that, when they see something beyond their programming, know when to await new directions. While some probes, such as the recently launched Deep Space-1, are a step towards greater autonomy with regard to decision making, we still have a long way to go before our tools decide what is interesting - on their own.

Robots are the most recent in a long succession of tools we have developed. For the most part we still control their actions directly - either by preprogramming a series of tasks - or by direct manipulation of their controls. When we use robots on Earth, in hazardous environments, for instance, the wait or "latency" between command and response doesn't pose a problem. With the aid of virtual reality and "haptic feedback" (using the sense of touch as well as sight), we can even "close the loop" and extend our reach through the eyes, ears and hands of our robots.

Many remotely operated systems now in place provide the clear, and productive illusion of telepresence. Recent developments in software and human/machine interfaces have allowed an ever increasing number of solutions to be available. Indeed, if the current pace of innovation is any guide, mission planners in the next decade will have many more options available to them as they consider the tradeoff between human and robotic paradigms. More importantly, these new tools will provide enhanced flexibility in blending the two approaches together in a synergistic fashion.

Closing the loop may be possible when using robots to explore hazardous terrestrial locations such as Chernobyl. We will also be able to operate robotic devices aboard orbital platforms such as the International Space Station from anywhere on Earth - or even on the Moon as humans did during the operation of the Russian Lunakhod rovers a light second or so away.

Alas, the not-so-virtual limits to virtual reality are reached when distance slams up against constraints imposed by the speed of light. Eventually the round trip time for signal and response overcomes the utility of direct human control of robotics. Such lag times would clearly rule out human control of robots on Mars from Earth, with a round trip light travel time of at least six minutes, even in the best of circumstances.

So, do these long feedback delays mean that remotely-operated vehicles are out of the question in the search for extraterrestrial life within our own solar system? And if so, are autonomously-operated vehicles the only choice by default?

The answer to both of these questions is clearly a resounding no! The important thing for all mission planners to remember is where the skills of human-based intelligence and machine-based intelligence best apply - and to be able to make the right decision when picking the appropriate technological approach. When long, one way travel times (physical and communications), simple defined tasks, and low payload weight are constraints, clearly autonomy is called for. However, when the tasks placed upon an exploration mission exceed the existing or projected capability of automation, or when the complexity (hence risk for failure) required for automation escalates to an unacceptable level, then another solution needs to be found. Humans can be part of that solution. So can robots. Let's look at what we might do on Mars as an example.

With humans on Mars in a central base camp, surrounded by highly capable robots, we could both extend our exploration to the entire planet as well as dramatically shorten our search for evidence of life. With this paradigm, the robots we bring with us may be thought of more as partners than designees. And capable partners they would be. Mars-based airplanes could give high-resolution views planet-wide, giving Mars-based humans ideas on where to concentrate the search. Real-time, human-controlled rovers could pick through the landscape, providing close-up imagery and samples for analysis. Intelligent "moles" could dig through the dust to sample primordial material, giving hints on past Martian climates.

When further analysis of something actually requires human proximity, then the decision is made as to whether to bring the sample to the base or to send a human out to the sample. If need be, we then hop in our Mars buggy and drive wherever we need. With all of these possibilities, and the scores of additional ones, the only way to truly explore for life is with humans in the loop - not just on Earth, but on Mars.

One of the clear lessons of our past 40 years in space is that humans will always lead the charge in exploration. Clearly, our tools are becoming more intelligent all the time, extending our reach into places that we'd scarcely dreamed of only a handful of years ago. But if we're serious about looking for life beyond our own home, our enthusiasm with these tools must be tempered by the knowledge that they are only extensions of ourselves. Tools, are, after all, only tools.

Humans have always used tools to extend their biological capabilities. Whenever a tool has been developed to visit places where humans cannot go unaided, be it space or the abyssal depths of the ocean, humans have always followed. Yet often times, tools such as robots are developed only after humans have already accomplished a task in person. Automated airplanes and remotely operated submersibles are obvious examples.

Oceanographic and Antarctic exploration provide two terrestrial research paradigms with clear applicability for the exploration of the solar system. These relatively mature activities embody a rich mix of forward-deployed humans and robots, surface ships and submersibles, autonomous in situ sensors and vehicles, and researchers working in their own home laboratories, all linked together by global satellite networks. In other words, the right tool for the right task in the right location.

There are clear synergies and overlaps between NASA's Human Exploration and Development of Space (HEDS) and Space Science Enterprises. Astrobiology, the search for life's origins and destiny, is at the core of this intersection.

The key to the successful and safe exploration of our solar system will be applying the proper balance of humans and robotic tools - not a bias toward one approach at the expense of the other. Both approaches have capabilities to offer the other. Not to employ this synergy would be to ignore the rich experience we have gained in pushing back frontiers and gaining knowledge on this planet.