Dr. Daniel Lee of the University of Pennsylvania presents the DARwIn-OP as a research and educational robot platform, sharing software approaches and examples before handing over to Stephen and Sunil.
Lee gives examples of Penn projects in outdoor navigation, quadruped locomotion, multi-robot search, and autonomous driving. LittleDog used planning and perception to choose footholds on rough terrain, while a team of robots used lidar and cameras to detect obstacles in an urban setting.
Lee describes an Urban Challenge vehicle handling a blocked four-way stop. Its MILE software waited up to 45 seconds before turning out of turn, then responded to another vehicle entering the wrong lane and passed the obstacle after a human driver backed up.
Lee recommends separating efficient, low-level C++ interfaces for sensors and actuators from higher-level scripting. Penn used MATLAB and Lua, with dynamically loadable libraries for speed-intensive work such as vision and kinematics. Restarting scripts lets students test changes without recompiling or rebooting the robot.
One approach runs MATLAB on the robot's Atom processor, with functions for motor control, camera images, and image processing. The alternative uses Lua, which Lee describes as compact and open source, with matching interfaces for the robot and Webots simulator. A Lua behavior uses camera images and inverse kinematics to make the simulated robot walk.
Lee shows how Webots combines a robot model, physics simulation, graphics, and a Lua terminal to run and debug behaviors and algorithms. He says DARwIn-OP can also run a graphical Linux environment on its Atom processor. Lee plans to make Penn's framework open source and invites users to build on it.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Michael Chang, a PhD researcher at the University of California, Berkeley, presents his work at the GRASP Lab Spring 2023 SFI on neural abstractions for modeling and controlling systems, and discusses future human-computer interfaces at DeepMind.
Karen Liu, an associate professor of computer science at Stanford University, discusses physics-based human motion simulation and its use in robotics at an MIT Robotics talk.
Dennis Hong, robotics professor and founding director at Virginia Tech, explains why humanoid robots suit human environments and how his lab developed soccer-playing robots and firefighting prototypes.
A ROBOTIS presenter discusses a Darwin humanoid robot that plays Dance Dance Revolution and the vision-processing and motion constraints behind the demonstration.
Alex, role and affiliation unknown, presents the Darwin Drives project with King at an ICRA 2012 challenge; they show a humanoid robot operating an iRobot Create and discuss possible uses for humanoid-driven vehicles.
Students from Purdue University describe their team’s Darwin-OP robotics project at an ICRA 2012 challenge, explaining its sensor-based reactions and demonstrating them.
