NASA is asking coders to create algorithms to improve 3D vision of its first humanoid robot in space for maintaining the International Space Station (ISS), freeing up astronauts for critical science and repair work.
Humans use glasses to help them see better, but for robots, the fix is in their code, NASA said.
The Robonaut Vision Tool Manipulation contest offers a total of $ 10,000 in prizes for the best algorithms.
Robonaut 2, or R2, is the first humanoid robot in space, currently being tested on the ISS.
Serving as an extra set of hands for station crew members, the robot is looking to help with the more mundane or repetitive tasks that are required for maintaining the million-pound laboratory, freeing up its human colleagues for critical science and repair work.
For example, R2 manages inventory using a Radio-frequency identification (RFID) reader and fastens bolts with a drill.
While astronauts can control R2 directly, making the robot more autonomous will make work on the station and on future deep space exploration missions more efficient. One goal is to help R2 “see” better, said NASA.
In order to use a tool, R2 relies on an algorithm to determine a 3-D representation of the tool. The algorithm works with the robot’s control system and allows R2 to create a plan for grasping objects and completing its tasks.
Existing algorithms assume that high-resolution images are always available, NASA said.
New algorithms are needed that can determine differences in objects based on noisy, stereo vision data.
The objective for the contest is to create algorithms that will receive a pair of noisy stereo images of common space tools such as an RFID reader, an EVA handrail, or a softbox, among others, and determine the correct 3-D representation of the object in the image pair.
Humans use glasses to help them see better, but for robots, the fix is in their code, NASA said.
The Robonaut Vision Tool Manipulation contest offers a total of $ 10,000 in prizes for the best algorithms.
Robonaut 2, or R2, is the first humanoid robot in space, currently being tested on the ISS.
Serving as an extra set of hands for station crew members, the robot is looking to help with the more mundane or repetitive tasks that are required for maintaining the million-pound laboratory, freeing up its human colleagues for critical science and repair work.
For example, R2 manages inventory using a Radio-frequency identification (RFID) reader and fastens bolts with a drill.
While astronauts can control R2 directly, making the robot more autonomous will make work on the station and on future deep space exploration missions more efficient. One goal is to help R2 “see” better, said NASA.
In order to use a tool, R2 relies on an algorithm to determine a 3-D representation of the tool. The algorithm works with the robot’s control system and allows R2 to create a plan for grasping objects and completing its tasks.
Existing algorithms assume that high-resolution images are always available, NASA said.
New algorithms are needed that can determine differences in objects based on noisy, stereo vision data.
The objective for the contest is to create algorithms that will receive a pair of noisy stereo images of common space tools such as an RFID reader, an EVA handrail, or a softbox, among others, and determine the correct 3-D representation of the object in the image pair.