As robots usually struggle with navigational concepts, a team of researchers from the Massachusetts Institute of Technology (MIT) have devised a new model to help them navigate surroundings more as humans do, reminding us the way hero Rajini Kanth in the film 'Robo' goes around dealing with the surroundings.
The new motion-planning model allows these robots determine how to reach a goal by exploring the environment, observing other agents, and exploiting what they've learned before enacting in similar situations. It teaches the robots to take into account what the other cars are going to do, as any human would do.
"Just like when playing chess, these decisions branch out until (they) find a good way to navigate. But unlike chess players, (they) explore what the future looks like without learning much about their environment and other agents," said co-author Andrei Barbu of MIT.
Even if they go through 1000 times the same crowd is as complicated as the first time, they keep exploring, rarely observing. Instead they now use stored knowledge as to what's happened in the past, Barbu explained.
MIT researchers have devised a way to help robots navigate environments more like humans do.
The new model combines a planning algorithm with a neural network that learns to recognise paths leading to the best outcome and uses that knowledge to guide the robot's movement in an environment os similar surroundings.
In an experiment, the researchers trained and tested the model in navigating environments with multiple moving agents, especially good for autonomous cars, especially navigating intersections and roundabouts.
In the simulation, several agents are circling an obstacle. A robot agent must successfully navigate around the other agents, avoid collisions, and reach a goal location, such as an exit on a roundabout.
The findings have been presented at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). The model can capture enough information about the future behaviour of the other agents (or cars) to cut off the process early, while still making good decisions in navigation.
With this model, planning gets more efficient and with additional training to the agents or cars to a few examples of roundabouts will help cars navigate safely, said scientists. "The plans the robots make take into account what the other cars are going to do, as any human would," Barbu said.
Going through intersections or roundabouts is one of the most challenging scenarios for autonomous cars. With this model, one day cars can learn how humans behave and how to adapt to driving in different environments, according to the researchers.
This is the focus of the Toyota-CSAIL Joint Research Center work. "Not everybody behaves the same way, but people are very stereotypical. There are people who are shy, people who are aggressive. The model recognizes that quickly and that's why it can plan efficiently," Barbu says.
More recently, the researchers have been applying this work to robots with manipulators that face similarly daunting challenges when reaching for objects in ever-changing environments, the way Rajini Kanth as a robot learned and navigated his way through to ward off detractors.(IANS)