By Daniel Vlasic, Rolf Adelsberger, Giovanni Vannucci, John Barnwell, Markus Gross, Wojciech Matusik, Jovan Popovi´c
Commercial motion-capture systems produce excellent in-studio reconstructions, but offer no comparable solution for acquisition in everyday environments. We present a system for acquiring motions almost anywhere. This wearable system gathers ultrasonic time-of-flight and inertial measurements with a set of inexpensive miniature sensors worn on the garment. After recording, the information is combined using an Extended Kalman Filter to reconstruct joint configurations of a body. Experimental results show that even motions that are traditionally difficult to acquire are recorded with ease within their natural settings. Although our prototype does not reliably recover the global transformation, we show that the resulting motions are visually similar to the original ones, and that the combined acoustic and inertial system reduces the drift commonly observed in purely inertial systems. Our final results suggest that this system could become a versatile input device for a variety of augmented-reality applications.
10. THE STANFORD CART
Grand Challenge finishers, UAVs, and even KITT from Knight Rider all owe a debt of gratitude to James Adams and Hans Moravec’s Stanford Cart. In 1979, the wagon traversed a chair-filled room on its own, a landmark achievement for self-navigating vehicles. Travel time: roughly five hours.
09. DANTE II
After eight volcano researchers were killed in two 1993 eruptions, robots were brought in to take the heat. The next year, Carnegie Mellon’s Dante II was lowered into Alaska’s steaming Mount Spurr to collect data. It fell in, but not before uploading its readings, making it the first “successful” terrestrial explorer robot.
08. DA VINCI SURGICAL SYSTEM
In the future, you’ll beg to be operated on by a machine. Credit Intuitive Surgical’s 2000 robot, a fusion of arms, cameras, and instruments that allows doctors to slice into patients remotely. Procedures done with the da Vinci are more precise than when humans wield the scalpel - research shows there’s less blood loss and quicker recovery.
07. THE MECHANICAL KNIGHT
Way back in 1495, Leonardo da Vinci designed what was probably the first robot - an automated suit of armor with a windup crank. It could sit up, wave its hands, and maybe even talk. Five hundred years later, engineer Mark Rosheim used the master’s schematics to build a working miniaturized version.
06. QRIO
Bipedal robots that can walk up stairs seem flatfooted compared with the running, jumping, and traditional-Japanese-fan-dancing Qrio. Officially, Sony uses its state-of-the-art androids, debuted in 2003, as corporate ambassadors. But the company may one day sell them for entertainment. Works for Beck: The singer recently used all six Qrios in his video for “Hell Yes.”
此乃Sony著名的QRIO機械人。QRIO源於英文“Quest for Curiosity”，意謂「好奇心探索」。QRIO身高58cm，體重7kg，在多達38個可轉動關節下，不僅可跳舞、唱歌、踢足球，更可以通過感測器判斷地面類型，並即時相應調整姿勢和步行姿態，來適應各種環境。QRIO還有更引人入勝的地方，就是能夠與人類進行豐富的交流。QRIO內置聲音與臉部特徵紀錄系統，只要對它說出自己的名字，它便會分析並記錄你的臉部和聲音特徵。到下一次見面時，它便會認出你，並向你問好。QRIO絕對是集科技與娛樂於一身的機械人。
05. SHAKEY
Developed by Stanford Research Institute International, Shakey had jerky, often nonsensical movements. But that didn’t stop the 1972 robot from entering the history books as the first machine to autonomously locate objects, steer around them - and then explain its logic for doing so.
於1966至1972期間由史丹福研究院（Stanford Research Institute，SRI）所研制，是全球首個懂得分析周圍環境並自行作下一項工作的機械人。其內置的電腦懂得處理簡單的移動、轉向和路線策劃；它亦可以組合一連串簡單的指令來執行較複雜的工作；它更懂得如何自行計劃和完成主人所指示的工作，亦可以將實行過的計劃儲存到日後再用。雖然Shakey偶爾會做出一些難以理解的行為，但它對人工智能和機械人的發展仍具有不少影響。
04. ROBBY THE ROBOT
Few robots can trace their origins to Shakespeare. Robby, from the 1956 film Forbidden Planet, was inspired by Ariel in The Tempest. But that didn’t keep Robby from leaving a legacy all his own. For decades, the very idea of a robot was synonymous with Robby’s bulbous figure.
03. SPIRIT AND OPPORTUNITY
Some robots sit in labs for researchers to tinker with. These two bots are on frickin’ Mars. Expected to last only three months when they touched down on the Red Planet in January 2004, the rovers are still going strong two years later - each sends back 100 megabits of data a day.
02. ASTROBOY While American kids were daydreaming of Superman, Japanese tykes were worshipping at the altar of Tetsuwan Atom, aka Astroboy. First drawn in 1951, Astroboy has rocket boots, lasers that shoot from his fingertips, and, uh, an ass cannon. The lovable crime-fighting robot was an inspiration to a generation of kids -some of whom went on to become robotics researchers. He’s a big reason why Japan is at the forefront of android development today. Domo arigato, Mr. Roboto.
01. STANLEY
The Stanford Racing Team’s autonomous vehicle is a modified Volkswagen Touareg that can scan any terrain and pick out a drivable course to a preset destination. Cup holders optional.
PuppetShow is a system that allows people use there bodies to control virtual characters. Puppet show uses a webcam to track the movements of a puppet and translates the gestures into bone movements in a game engine. Puppet Show’s goal is remove confusing middleware (keyboards and mice) from the player interface and allow the users to use their body’s natural expressiveness to convey themselves in virtual environments.
PuppetShow was built with Processing and runs within the Unreal Runtime using custom characters. Currently PuppetShow is in the late alpha phases and can track any flat color puppet and relay the actions two different specialized characters: a duck and a panda. Currently the PuppetShow team is integrating the complete PuppetShow animation system into the standard Unreal skeletal system. In the near future PuppetShow will be able to control any Unreal character through the PuppetShow interface.
How it Works
PuppetShow in truth is actually two separate systems that communicate over a TCP link: the puppet tracker and the Unreal puppet animation system. Below is the overall flow diagram of how PuppetShow works.
Puppet tracking is accomplished by using a puppet, a webcam, and the PuppetShow color tracking software. The color tracking software uses a refined blob tracking algorithm to track solid colors on a puppet. Since the tracking is based solely on color almost anything can act as a puppet: origami, gloves, socks, shirts and so forth! The tracking color can be set easily with the dynamic color calibration system. Based on a predetermined scheme, the tracking software then takes the blobs and translates them into usable information which is subsequently sent to the Unreal Runtime running the PuppetShow animation system.
The PuppetShow Mutator takes in values from the color tracker and translates them into bone translations and rotations. These are then applied to the character controlled 3D mesh (in the above diagram: a ducky!) which is subsequently displayed on the screen. An important note: PuppetShow does not interfere with the standard input devices of Unreal. This allows players to move and manipulate their characters as they would usually be able to.
The Muppet Mobile Lab is a theme park exhibit for Disney’s California Adventure. Part of Disney’s “Living Characters Initiative” project, Dr. Bunsen Honeydew and Beaker are “brought to life” through the use of advanced Audio-Animatronic remote-control-puppeteering technology.
BVH – skeletal data
;){kind=link}