Learning by Demonstration

It’s never easy to program a robot to perform a useful task. A person has to first attend a course or pick up some skills from a robotics manufacturer. There’s always a designated person in the factory that is able to program the robot and no one else. This is not true for a human operator as all we have to do is instruct the person to perform the task, and this is possible if you know the task to be accomplished.

Many researchers have worked on Learning by Demonstration and the following video is the work done at Willow Garage where they run trials on normal end-users. The end-users are able to physical move the PR2 to show it how to perform a particular task engaging the robot using speech recognition. This is one of the key things that will bring robots closer to a human centric environment.

Autonomous Golf Cart from NUS

National University of Singapore(NUS) is working on developing autonomous vehicles that can be used for urban systems of the future. The aim of the project is to achieve autonomous navigation with the lowest cost allowable without sacrificing the robustness and reliability of the system.

This is autonomous golf cart equipped with 3 LIDAR on the vehicle for feature detection, curb detection and obstacle avoidance. Other sensors used on the cart are is the camera and the odometry system. This allows the cart to localize it’s position and navigation autonomously. The video shows the cart has the ability to perform SLAM and handle human traffic in a low traffic place like the university.

The Google autonomous car has set very high standards for autonomous navigation with features within the system to plan it’s actions at road junctions and real life traffic conditions. This might come at a high cost given the sensors that are used on the google car. NUS hopes to lower the cost of development. There are uses for autonomous vehicles within pretty controlled environments like university campuses and small towns.

Open Source Robotics Foundation (OSRF)

Willow Garage has been spearheading the push of open source software for use in the robotics community. Robot Operating System (ROS) has been used widely in the research community and it’s interesting how researcher can share their work globally and how this community not only helps to grow the functions of robots but also improve the robustness of the algorithms.

Open Source Robotics Foundation (OSRF) was formed as a bigger umbrella to take care projects like ROS and Gazebo to unify the use of the platforms and also to educate the users of the platform. These evangelistic or consultants will preach the good news and they will be the eyes and ears on the ground to find out the needs of robotics as a whole globally.

Kinect Guru’s New Robot Project

Garratt Gallagher has done many hacks for the Kinect. He won first place in the Kinect/ROS competition. He’s started a KickStarter project – Bilibot. It’s a iRobot Create with the Kinect as a sensor for navigating the robot around. ┬áKinect is a great tool and there are still competitors on the market hopeful to get a bite of the pie. They have changed their business model from selling products to being OEM for consumer product manufacturers. There’s a lot of prospect in this area given how microsoft has brought down the cost to this level. The rules of competition is different and the consumers are there to reap the rewards of this battle. Can’t wait.

The new Kinect will hopefully push cheap robots to a higher level of competence.

Hizook has covered this story in detail.