Incremental Blogger

I’m going to take a break from my regular topics and blog about something that’s been a long term interest of mine: Spatial and Visual Searching.

“Huh,” you say? What’s “Spatial and Visual Searching?” Well, there’s no real definition, it’s just a phrase I made up back in school when I was working on autonomous Robots. Basically, what I’m referring to is the problem of an autonomous Robot (or similar) spatially searching a space using a vision system as a primary sensor (though this isn’t exactly necessary, but yet given the state of sensing systems now and then, it’s a reasonable assumption that this will be a Robot’s primary searching sensor).

My particular interest was in a Robot searching indoor spaces, such as in a house or office building. I’m most keen on indoor Robotics because after all that’s where many of us spend much of our time and it seems natural that a Robot would too. Indoor environments also are more of a constrained environment than let’s say searching a mountainside or possibly even a lake bed. In fact, I imagine a whole additional set of challenges and discussions could center around searching unrestricted spaces, such as outdoor venues, multiple buildings, forests, you name it. Anyway, I was interested in how an autonomous Robot might try to find something–let’s say a bucket that it was asked to retrieve or the whereabouts of a person.

[Anyway, I'm not sure how far I'll go with this post today, but I'm thinking at least I'll get it started and then possibly add more to it at a time. Most blog posts aren't indended to be added to over time, however, this one most likely will be. I'm going to try to record some of my thoughts on the problems, in part to archive it and in part to possibly spur some insights. Well, here goes.]

Before I get too deep, I’ll lay out a couple of boundaries to my thinking:

For now, this discussion will be geared around a physical Robot searching for something as it slides across a single-level floor, with a known floorplan. There’s no doubt that there are many other possibilities. For instance, back in the day, I used a gantry Robot when I initially tried tackling this problem. It was a conventient way of navigating across a space with minimal problems, but it sure isn’t a configuration that would be likely in a home. I’m also keen on virtual Robots, but that’s something for another post.

Vision will also play a lead role in how I envision a Robot searching for something indoors. That doesn’t mean that other sensor aren’t useful–it’s just that a vision system is a good, general purpose sensor that’s fairly practical to use.

With these boundaries in place, let’s start at the beginning: What do I mean by “spactial searching” for Robots? Well, it’s really no different at a conceptual level than what you have to go through to find something in your home. If someone asks you to find their glasses, how do you go about doing it?

You might start by thinking in terms of how easy it might be to see what you’re looking for. You might also consider where the object might be–with the primary objective here I presume to minimize the time and energy it takes to locate the object. From this list of possibilities, you might prioritize it and create a series of plans on how to search this space. These search plans might be hierarchical or exhaustive or who knows. It might depend on what you are searching for and how easy it is for you to use your sensing abilities to locate it. In other words, if it’s a small, translucent object you might have to be very close to it to see it or you might need to be careful when there’s not enough light or enough contrast or too much light or, or, or. Much depends on the sensors used and their capabilities, if you will. Don’t think object oriented, think sensor oriented.

Ooops….I’m being beckoned. I didn’t get very far….but I’ve got to run. This is chemo week so I’ll do what I can to get back ASAP….So much for the best laid plans…lol.

David Pogue’s videos sure have gotten more interesting–in part because of their better production. The little details matter. He’s quite entertaining.

When I watch his clips it increases my belief that there need to be even better tools for producing nicer videos. I ask myself, “What technologies do you need to produce videos similar to how he does it?” or “What two or three techniques might make single-person produced even better?”

Where are Robots–or at least automated cameras–when you need them?

Disney bots to be shown at Maker Faire.

This year (uh, I mean last year) in Robots.

Robert Scoble blogs about taking pictures of an autonomous helicopter being flown at Stanford. Very cool!

How did I learn about Robert’s post? Twitter via GTalk. (I set up Twitter to forward tweets to GTalk the other day.) What’s really cool is that GTalk even provides an instant preview of Robert’s posted photos on Flickr. This is a much better set up than watching Twitter’s own web pages. No photo preview there. I’m guessing that other twitter-supporting apps do the same thing? I guess I’m behind the times. Again.

Dave, This gives an idea of what Twitter might be like if it supported thumbnails.

The GMC Robot ad was my favorite Super Bowl XLI commercial. What can I say? I like Robots.

By the way, it was hard finding a list of the Super Bowl commercials online. I remember a link being mentioned on TV, but I could remember the URL. So I tried searching for “Super Bowl commercials” on Google, however, all the top links were for previous Super Bowl events. The official Super Bowl site gets returned too, but again, after a quick scan I didn’t see a link to the commercials. I presume I missed it.

I finally turned to searching the blogs (using blogsearch.google.com) and sure enough in a few minutes I had a link to a partial list of ads on videos.google.com and a more complete list at AOL. I’m fairly sure this isn’t the link promoted on CBS, which is maybe a lesson to the folks wanting to get their URL out. (I also checked news.google.com, however, for the handful of online articles I read, no one appeared to provide the link to the commercials themselves. They just talked about the commercials.)

This all illustrates how difficult it can be to find current information online. The “best” links are often based on criteria built up over time, which is something that a breaking event doesn’t have in its favor. Of course, ask a person for a link of Super Bowl ads this morning and they’d know exactly what you are looking for. They aren’t going to be handing me a link to 2004 ads–sorry Google, wrong answer.

Fortunately searching blogs got me around this predicament. It’s a trick I often use when looking for breaking “news.” It worked this time, however, it sure would be nice if there was a search engine that was biased towards new information on the Net.

CNet is reporting that Microsoft will officially launch its Microsoft Robotics Studio. “The software platform includes a 3D tool to simulate robotics applications; a services-oriented runtime that lets applications communicate with a wide variety of hardware; and a visual programming language that lets nonprogrammers easily program robots with a drag and drop icons.”

This is great to hear. Next up I’d like to see Microsoft launch a multi-camera and multi-image stream vision SDK. Intel has done quite well with its OpenCV vision SDK, however, there needs to be something that has a managed code SDK, simply because of all of the managed developers in the Windows world and ease of access to developers.

Why a vision SDK? First, more and more computers and mobile phones have cameras built right into them. Why not provide an SDK to work with their image streams? Already Microsoft provides imaging apps with their webcams, so launching an SDK seems like a great next step and will open up a whole new category of developer support not only for standard computers but also potentially for Robotics efforts, home security, interactive art projects, advanced cameras, and the like.

Why do I love reading Make? Because of projects like this: An RC car mofied to sweep the floor.

Microsoft is getting into Robotics with Microsoft Robotics Studio CTP 1. [Link via Robert Scoble]

It’s great to see Microsoft join this active community of enthusiasts. I just downloaded the CTP, but unfortunately I discovered too late that it requires VS 2005, so I’ll have to try it again later on my other system. This one only has VS 2003.

The studio introduces a runtime environment that is geared towards interacting and controlling multiple devices and sensors at once. From the Robotics site:

“Microsoft Robotics Studio Runtime provides a service oriented architecture which combines key aspects of traditional Web-based architecture with pieces from Web Services to provide a highly flexible and lightweight, distributed application model.”

Nicely done.

The Studio also includes services from some existing third-party hardware and a simulator. I guess the simulator–if it can be used for visualization–might be interesting, but a simulator in itself misses all the “fun” stuff in Robotics: the noisy data, reflections, slippage, and so on. The real challenge is in the real world. But it’s nice to have it anyway. Like I mentioned earlier I haven’t successfully installed the Studio yet, so maybe it supports visualizing sensed data and the like. I look forward to checking it out.

The Robotics Group manager, Tandy Trower, posts his team’s philosophy here.

“Robotics has long been a technology area that has captured the attention and expectations of many. We think robotics is poised to take off rapidly, and there are solid indications that this is true! With component hardware costs coming down and computational capabilities increasing, the robotics industry appears to have the right conditions to really grow quickly.”

The Robotics community has never been lacking for optimism. In fact, there have been many such statements over the years. However, things are different today. There is progress. And most importantly things are becoming more practical. That’s the key.

By the way, the Robotics team even has a blog. Check it out here.

A Channel9 interview is here..

Update: I got the Robotics Studio CTP installed and played with the Robotics Simulation Visualizer a bit. It took me a few minutes to read through things and figure out what to do. The first step is to launch the Robotics Studio Command Prompt. From it you can type the statement:

bin\dsshost /p:50000 /manifest:”config\SimulationTutorial1.xml”

This will cause the simuilation window to appear. You can also browse the services involved using your browser by navigating to the URI: http://localhost:50000.

Now that Sony’s Aibo Robotic dog project has been dissassembled, it looks like Dasa Tech of Korea is going to save the day by introducing their own Robotic pet dog. [Link found via Memeorandum]

Frm the Dasatech website, some of the dog’s features are:

* 334×300x192mm dimensions
* Sensors: Temperature, PIR, tilt, touch, infrared, PSD, microphone
* Vision: CMOS color camera (355×288) in nose
* Processor: 32bit RISC, 8bit CPUx11
* Lithium batter, 1.5hrs battery life in “Auto Mode”
* 17 actuators
* Display: 3 color LED dot matrix (31 dots, 7×5)
* Sound: mono, music download support
* Voice recognition: 30 words
* Bluetooth, serial cable

No price posted yet.

I wonder if we’ll be seeing these robot dogs at holiday season.

It doesn’t sound like there is much programmability available here. I bet with some creative thinking, someone will come up with some way to extend and modify these Robot though. In fact, wouldn’t a UMPC and a pet Robot like this make an excellent combination?

After two years the Mars Rovers are still roving the Martian landscape. It’s simply amazing.

The original expectation was for the Mars Rovers to last 60 or possibly 90 days. So much for predictions. Two years later the Rovers are still chugging along. And there are some new predictions that the Rovers could last another year or possibly two. Who knows.

What this all points to though is a slow transition both in space exploration and even here on Earth to remotely guided machines in more and more cases. I recall awhile back that there was great concern whether mechanized explorers were the right way to go. Shouldn’t we send people instead? The answer is “Yes,” to both. There will be times when mechanized, teleoperated, or semi-autonomous vehicles will be ideal and there will be times when people will do better. A world with one or the other is not the solution. Predictions of such are more and more looking like poor bets. I’m guessing the odds will continue to grow against them.

Stanford’s Stanley Robotic vehicle has travelled 136 miles autonomously to complete Darpa’s Grand Challenge.

Quite amazingly three vehicles have made it across the finish line this year. Last year (the first year) no Robot travelled more than eight miles.

Congrats to the teams!

Here are some videos:

The starting line.

Here’s a video of the Stanley Robot among others making a 90 degree turn on a dirt road. (The second vehicle (blue Volkswagen Touareg R5) in the video is the winning Robot.)

Video of several Robots crossing a paved bridge.

This holiday weekend Mars rovers Spirit and Opportunity will still be canvasing the martian surface. It’s simply amazing that the Robots have kept on ticking–despite all the little challenges. I see that Opportunity was stuck in a sand dune for the last five weeks, for instance.

Tonight’s Nightline episode was dedicated to the local Carl Hayden High School Robotics group and their underwater ROV competition win against several universities including the likes of MIT. It’s a true underdog story.

The story is particularly interesting in light of Scoble’s and others’ comments on inspiring the current generation of students to pursue engineering and science fields.

There’s more on the Falcon Robotics team here in a Wired magazine article. It’s the stuff of movies. A must read.

I met team-member Lorenzo, their co-teacher Allan Cameron, and several others on the team (sorry I can’t recall their names) at the recent Microcomputers in Education Conference at ASU. The high school Robotic team had a booth near the Tablet PC table I was manning. In downtime the kids came over to experiment with the Tablets on display and I snuck over to check out their Robots.

They were all interested in shaking down the Tablets to see what they could do. In no time at all Lorenzo had Physics Illustrator hurling objects through a maze of springs and ropes. And others were painting in ArtRage, flooding Journal with ink, and challenging MathJournal to solve random equations. It was terrific to see them contemplate the possibilities.

It didn’t take long before they had ideas about how they could use a Tablet to their advantage in future competitions. I only wish I had an extra Tablet to donate to their efforts. Their energy is infectious.

Congrats to the Falcon team. It’s great to see such creativity. Keep thinking big.

All robo-wheels lead to the RoboBusiness conference being held this week now in Cambridge, MA.

Here’s some robo-news from the conference:

iRobot Targets Roomba Knockoff

Boeing Sponsors RoboBusiness Conference & Expo

Robotics companies merge, aim at high-end business use

Robots: Research in motion

Robot makers converge at confab