Development of Autonomy for Log-loading Machines in Canadian Mill Yards

The timber-harvesting and processing industries anticipate that the use of robotics and AI technologies will increase
productivity, safety and job satisfaction of machine operators to help address the shrinking labor force. We have engaged
with our partner, FPInnovations, on research to increase the autonomy of log-loading machines with the goal to demonstrate
the technology on machines operating in the mill yard. We will focus our research and development on three relevant
problems, of direct relevance to different aspects of log-loading operations. In the first project, we will investigate the use of
vision and inertial sensors on the crane of the machine to determine the position of the crane’s grapple and to measure the
crane’s joint angles. In our second project, we will develop localization and mapping solutions to allow the machine to
position itself relative to the infeed deck of the mill and to build a map of the infeed deck area in order to increase the
machine’s situational awareness. In the third project, we will work towards an integrated planning and control framework for
efficient unloading of logs from large piles in mill-yard storage areas. To validate the developed algorithms, we will employ
the physical Crane test-bed available at FPInnovatrions, as well as, state-of-the-art simulation tools.

Faculty Supervisor:

Inna Sharf

Student:

Partner:

FPInnovations (Pointe-Claire, QC)

Discipline:

Engineering

Sector:

Agriculture; Manufacturing; Professional, scientific and technical services

University:

McGill University

Program: