Listen to this article
How fresh is your data? For drones Searching a disaster zone or robots inspecting a building, working with the freshest data is key to finding a survivor or reporting a potential hazard. But when many robots simultaneously transmit time-sensitive information over a wireless network, a data traffic jam can occur. Any information that comes through is too out of date to be considered a useful, real-time report.
Now, MIT engineers may have a solution. They developed a method to adapt any wireless network to handle a high load of time-sensitive data coming from multiple sources. Their new strategy, called WiSwarmconfigures a wireless network to control the flow of information from multiple sources while ensuring that the network delivers the freshest data.
The team used their method to tweak a conventional Wi-Fi router, and showed that the adapted network can act as an efficient traffic policeman, able to prioritize and relay the freshest data to keep the many vehicle tracking drones at work.
The team’s technique, which they will present in May at the IEEE’s International Conference on Computer Communications (INFOCOM), offers a practical way for multiple robots to communicate over available Wi-Fi networks so they don’t have to have to carry large and expensive communications and processing. hardware onboard.
Last in line
The team’s approach departs from the standard way in which robots are designed to communicate data.
“What happens in most common networking protocols is a first come, first served approach,” said MIT author Vishrant Tripathi. “A video frame comes in, you’re processing it. Something else is coming in, you’re processing it. But if your task is time sensitive, like trying to detect where a moving object is, then all those old video frames are useless. What you want is the latest video frame.”
In theory, an alternative approach of “last in, first out” could help keep the data fresh. The concept is similar to a chef putting out entreés one by one as they are hot off the line. If you want the freshest plate, you want to be the last one in line. The same goes for data, if what you care about is the “information age,” or the most up-to-date data.
“Information-age is a new measure for the freshness of information that considers latency from an application perspective,” said Eytan Modiano of the Laboratory for Information and Decision Systems (LIDS). “For example, the freshness of information is important for an autonomous vehicle that relies on various sensor inputs. A sensor that measures proximity to obstacles to avoid a collision needs fresher information than a sensor that measures fuel levels.”
The team looked to prioritize information-age, by incorporating a “last in, first out” protocol for multiple robots collaborating on time-sensitive tasks. They aim to do this with conventional wireless networks, as Wi-Fi is widespread and does not require massive onboard communications hardware to access.
However, wireless networks have a major drawback: They are distributed in nature and do not prioritize receiving data from any one source. A wireless channel can quickly become clogged when multiple sources are simultaneously transmitting data. Even with a “last in, first out” protocol, data collisions will occur. In a time-sensitive exercise, the system will crash.
As a solution, the team developed WiSwarm — a scheduling algorithm that can be run on a centralized computer and paired with any wireless network to manage multiple data streams and prioritize the freshest data.
Instead of trying to retrieve every data packet from every source at every moment in time, the algorithm determines which source in a network should send data next. That source (a drone or robot) will observe a “last in, first out” protocol to send their freshest piece of data over a wireless network to a central processor.
The algorithm determines which source to relay data to next by assessing three parameters: overall weight of the drone, or priority (for example, a drone tracking a fast car may need to update more often, and therefore will have a higher priority than drone tracking. slower vehicles); age of the drone’s information, or how long it has been since the drone sent an update; and drone channel reliability, or probability of successfully transmitting data.
By multiplying these three parameters for each drone at any given time, the algorithm can schedule drones to report updates via a wireless network one at a time, without blocking in the system, and in such a way as to provide the freshest data for the successful execution of an hour. -sensitive work.
The team tested their algorithm using multiple mobility-tracking drones. They equipped the flying drones with a small camera and a basic Wi-Fi-enabled computer chip, which it used to continuously relay images to a central computer instead of using bulky, onboard computing system. They program the drones to fly and follow small vehicles that move randomly on the ground.
When the team paired the network with its algorithm, the computer received the freshest images from the most relevant drones, which it used to send commands back to the drones to keep them on the vehicle’s track.
When the researchers ran experiments using two drones, the method was able to relay data that was twice as recent, resulting in six times better tracking, compared to when two drones performed the same task. experiment with Wi-Fi only. When they expanded the system to five drones and five ground vehicles, Wi-Fi alone couldn’t handle the heavier data traffic, and the drones quickly lost track of the ground vehicles. With WiSwarm, the network is better equipped and enables all drones to continuously track their respective vehicles.
“Ours is the first work to show that the information-age can work for real robotics applications,” said MIT author Ezra Tal.
In the near future, cheap and agile drones may work together and communicate via wireless networks to perform tasks such as inspecting buildings, agricultural fields, and wind and solar farms. In the future, he sees the approach as essential for managing data streaming throughout the smart city.
“Imagine self-driving cars coming to an intersection with a sensor that detects something in the vicinity,” said MIT’s Sertac Karaman. “Which car should get that data first? This is a problem where the timing and freshness of the data is important.”
Editor’s Note: This article was republished from News at MIT.