Established in 2007, the RIM Lab at CMU is focused on research in advanced robotics such as mobile robots, auditory mapping, teleoperation and human-robot interaction.
ET Building 119
Brian P. DeJong
Omnidirectional Spherical Robot
||Spherical robots have several advantages in dangerous environments - they can be incredibly mobile and agile, durable and sealed off from the environment. Many spherical robots change their internal center of mass (CoM) to roll around, but most are steered like a wheel rather than omnidirectional - they must roll forward or backward to turn. |
We have designed and built a novel omnidirectional remote-controlled spherical robot that uses six pendulums to change its CoM. The design has both directionality and mobility advantages over previous CoM designs.
[Holliday, C., Quintal, D., Duthie, M., Asiri, A., DeJong, B.P. (2012). Omnidirectional RC Robot. ASEE, North Central Conference.]
Auditory Occupancy Grids on a Mobile Robot
The auditory environment around a mobile robot can be important information - a robot can better follow verbal commands, recognize suspicious sounds or find survivors in rubble if it has a map of the sounds around its workspace.
Localizing sound is a challenge on mobile robots because their microphones are closely positioned, but this can be overcome by the robot's mobility. By fusing multiple measurements around the room, a mobile robot can create an accurate map.
We have simulated and implemented a probability map of sound sources called an Auditory Occupancy Grid (AOG) in both two and three dimensions. The map successfully locates multiple sound sources using only a few measurements.
[1. DeJong, B.P. (2011). Three-Dimensional Auditory Occupancy Grids: Accuracy and Robustness. Pittsburgh, PA: IASTED International Conference on Robotics.
2. DeJong, B.P. (2010). Auditory Occupancy Grids: Sound Localization on a Mobile Robot. Cambridge, MA: IASTED International Conference on Robotics and Automation.
3. Simon, D.C., DeJong, B.P. (2010). Robot Sound Source Localization Using a Host Computer. ASEE, North Central Conference.]
Lifelike Robotic Arm
We have designed and built a seven degree-of-freedom (DoF) lifelike robotic arm for teleoperation applications. The arm uses seven servomotors to actuate the shoulder (three DoFs), elbow (one DoF), wrist (two DoFs) and a pressure-sensitive gripper (one DoF). The arm is dexterous and strong enough to lift a ceramic cup, yet gentle enough to not crush a paper cup.
[1. Crosswait, C., Roell, S., Stack, N., Tice, J., DeJong, B.P. (2010). A Seven Degree of Freedom Lifelike Robotic Arm, ASEE North Central Conference.
2. Carter, C., Dwyer, B., Lutsic, R., Mills, S., DeJong, B.P. (2009). Wireless User Controlled Robotic Arm. Grand Rapids, MI: American Society for Engineering Education's North Central Sectional Conference.]
Mobile Star Finder
The Mobile Star Finder (MSF) helps amateur astronomers by pointing out stars of interest using a visible-light laser. The user enters their location, time and celestial body of interest, and the MSF uses its two degrees-of-freedom to locate the body in the night sky.
[Duthie, N., DeJong, B.P. (2010). Mobile Star Finder, ASEE North Central Conference.]
Accelerometer-Driven RC Car
Instead of a traditional RC controller, the accelerometer-driven RC car (ADRC) uses an accelerometer in its transmitter to control the RC car. The user tilts the transmitter, and the car moves in the corresponding direction.
[Prawdzik, N., DeJong, B.P. (2010). Accelerometer-Driven RC Car. ASEE North Central Conference.]