In September 2012, MIT unveiled its prototype Simultaneous Localization and Mapping (SLAM) equipment. The device is a real time mapping system that uses Microsoft Kinect, a laser range finder and a laptop. Kinect is a motion sensing input device by Microsoft for the Xbox 360 video game console and Windows PCs. It is based around a webcam type peripheral for the Xbox 360 console.
The SLAM device was designed to be used by first responders at emergency sites. Unlike most of the other real time mapping systems, this does not rely on a robot or any other external sources. The unit is worn by a person. There is a backpack which contains the battery source and a laptop. The portion that is strapped to the person’s chest contains a Microsoft Kinect camera at the top and a laser rangefinder on the bottom.
The sensors work by scanning a building in a 270 degree arc with a Light Detection and Ranging (LIDAR) laser. In combination with the Microsoft Kinect, depth and visual data is detailed and a map is created. To take into account the wearer’s gait and compensate for tilting, a Microstrain inertial sensor is used. This is a device that is designed to provide three-axis static and quasi-dynamic orientation measurements. This sensor allows the system to compensate for how a person walks. A robotic device has a level and predictable way of moving, since a person does not normally have a steady gait, especially in an emergency situation, the sensor compensates for this allowing a correct map to be created.
The Kinect’s camera can tell if it has seen a room or hallway before, the technology is designed so that when a person walks through the same space again retracing their steps, images are compared and the map is updated with more accurate information. There is also a barometric pressure sensor which allows it to map multiple floors while keeping track of each one.
Research scientist for MIT’s Computer Science and Artificial Intelligence Laboratory, Maurice Fallon is the SLAM project leader. He foresees the unit being used in hazmat type situations where the first responders would be wearing a full suit to go in and explore an environment. Fallon says, "The current approach would be to textually summarize what they had seen afterward – 'I went into this room on the left, I saw this, I went into the next room,' and so on. We want to try to automate that." You can see this in action by viewing a video created by Fallon.
The original prototype unit has a button that can be used to annotate the map. The developers see the progression as being a system where you can add voice or text tags to the map. This would allow the system to record especially dangerous areas so they can then be marked on the map as the person walks through.
While the SLAM unit does have a laptop in the backpack unit, the information can also be transmitted wirelessly to an off-site computer. Researchers believe that the SLAM device can be miniaturized to make it better suited for emergency responders. The researchers say, "We envisage that the ?nal device will be a hand-held unit, similar in size to a miner's lamp or... installed on the shoulder of the user."
This project was funded by the US Air Force and the Office of Naval Research. In October 2012, the researchers delivered a white paper at the Intelligent Robots and Systems 2012 conference in Portugal. They said, "Our work is motivated by rapid response missions by emergency personnel, in which the capability for one or more people to rapidly map a complex indoor environment is essential for public safety."
Edited by Brooke Neuman