Hybrid human/robot coordinated joint motion control architecture for application in unstructured environments involving iterative or repetitive tasks

Case ID:
UNR02-015

Robotics used in the manufacturing industry cannot be aptly applied in machinery operations that require complex operations and variable tasks or a changing environment that is unstructured and requires ongoing adjustment.

The University of Nevada, Reno's Coordinated Joint Motion Control (JMC) architecture allows for artificial intelligence enhanced computerized control of a variety of machines from small robotics to heavy equipment such as large excavators. Using the novel JMC architecture, the human operator retains the master control of the main motion parameters, while the computer coordinates the repetitive/iterative movement. The operator can make adjustments to the motion in real-time, as needed, to accommodate rapidly changing environmental conditions.

The JMC architecture also includes error correction mechanisms to compensate for the mechanical inefficiencies that may arise from imperfect mechanical sensors, actuators, and controls in hydraulics and articulated heavy machinery. These mechanisms effectively map the digital domain to real-world applications.

The JMC architecture supports joint-link hydraulic controls, electrical robotic manipulators, or even virtually controlled devices. It can be used with electrical, hydraulic, or pneumatically powered machinery.

The architecture improves loading efficiency and cycle times reducing energy requirements. Unburdening repetitive tasks from the human operator helps reduce fatigue and improves safety. Ease of use of the equipment is improved and as a result operator training and learning time is reduced. A working prototype has been developed using a Bobcat 435 excavator. Preliminary field trials have shown:

17% reduction in the number of bucket loads
26% reduction in fuel consumption
39% reduction in loading time

Potential Applications

  • Material processing, Earthmoving and Mining
  • Loading and Packing
  • Construction

Opportunity

Available for licensing. UNR is seeking expressions of interest from parties interested in collaborative research to further develop, evaluate, or commercialize this technology.

Patent

Patent Information:
For Information, Contact:
Cara Baird
Licensing Associate
University of Nevada, Reno
carab@unr.edu
Inventors:
George Danko
Keywords:
Robotics