Tree-Swinging Robot

Forestry is a major export industry, but workers face some of the highest risks of injury and fatality. Felling trees is potentially dangerous, and the terrain is often steep and difficult.

To provide a long term solution, students and scientists at the University of Canterbury, along with with Scion scientists, have designed, built and developed a tree-traversing robot that can move through plantation forests by gripping trees rather than the usual wheeled vehicles that also cause soil damage.

The fully functional remote-controlled tree-traversing robot, at a quarter of the final size, is the first of its kind in the world.

The machine will be able to cut trees down remotely without the operator being close by. And it will be able to move around the forest without concerns about steepness, soil damage or the weather.

The trunk-gripping mechanism (think of it as two hands) allows the robot to rotate around the tree and to accurately grasp any other trunk within its maximum reach in any direction.

The prototype includes actuator control translated into joy-stick operation by the forester.

In a separate research project, an innovative cutting mechanism was designed, built and added onto the existing robot.

When the current quarter-scale design is scaled up to full size the resulting machine will be heavy, so the aim of this work is to develop and produce a lightweight machine which retains the ability to cut down standing Pinus radiata trees and keep the operator a safe distance from the tree.

The project was dreamed up 10 years ago by Richard Parker, a senior scientist at Scion.

Dr. Parker specialises in forest harvesting and rural fire fighting, as well as ergonomics and human factors in dangerous and demanding tasks. He is also involved in forest teleoperation and robotics research, aimed at removing workers from high risk situations.

He had a couple of years as a logger and he is a volunteer fire fighter. “I know what it is like to be on the end of a chainsaw,” he says.

He also developed high-vis shirts with yellow stripes in the early 1990s as a research project for the industry.

“From working in the bush and being in the bush as a researcher I could see if you had a machine that could use trees for support, it would be an advantage, because the ground is so complicated, with rocks, creeks and weeds.”

“Animals like apes or snakes get around a forest from tree to tree. Way back in the past my first qualification was in zoology, and there are types of snakes which jump from tree to tree. Why can’t we have a machine that does the same thing?”

Until he moved to Christchurch four years ago, the idea stayed in his mind, but working at Scion, which is on the University campus, he rubbed shoulders with engineers who can make things.

“I am not an engineer and I built a model with balsa wood and syringes acting for the hydraulic rams. Then later I built a second “Heath Robinson” model with carbon fibre and wire.   That was the model I used to show the University’s robot professor.  She was able to take the ideas and guide a group of final year engineering students to build the machine.”

Dr Stefanie Gutschmidt, who is a senior lecturer in mechanical engineering helped a group of engineering honours students in developing a working prototype. The students had nine months in which to deliver their project as part of their engineering honours degree.

Two of the four students focused on the design and two on the mechatronics aspects of the project.

She says that the research and development projects at the University of Canterbury are real life industry projects, that they bring people from different disciplines together to work on the project from an idea right through to the final product.

Dr Parker says: “She has a really good knowledge of engineering and what works and what doesn’t. She was key to getting the working prototype going, and along with her we have a PhD student Chris Meaclem who is doing all the control work on the machine.”

Chris has the machine working semi-autonomously, where he points to the tree it will work on next.

Recently another group of students have designed, built and added a chainsaw cutting unit to the initial robot, and both are in a cage so they can be operated and tested safely. In time the robot will have two chainsaws, one on each arm.

The quarter-scale prototype weighs 50kg and has a 2.5m reach, so the trees on which the robot is tested are quite close together.

“Initially we were thinking about a reach of 8m, so it would be a pretty big machine, but we are going to cut our teeth on a smaller machine with a 4m reach first.”

Stefanie says there are many areas of operation for this innovation besides cutting down trees, eg thinning to waste, pruning, monitoring and measuring. Currently, however, the prototype is optimised in view of the best strength-to-weight ratio. The team is very positive about the outcomes.

At the moment the robot is built from aluminium plates, and to make the machine larger it will have to be lighter and stiffer and will also cost more.

“The main criteria of our current optimisation efforts is to have a fully functional, reliably working machine, and only later we will of course include other criteria such as the cost factor. We want to find the best solution first.”

“The project is certainly an exciting design and innovation project. It’s such a motivator for the students when we are able to offer “from scratch” engineering problems.

“We allow students to think outside the box and be innovative themselves. In this regard we are not constraining them in any way, although we do have to work within a given budget. A full-scale machine is aimed at a reach of 8m, which will require quite a different design concept altogether. For instance, a different power supply and actuation concept (hydraulics) needs to be designed into the existing concept.

The project has funding through the Primary Growth Partnership with Future Forests Research.

“The Ministry of Primary Industry has seen the vision and could see we needed a completely revolutionary approach. They put up half the money and the industry including forestry companies and some contractors, put up the other half.   They are all taking a great deal of interest in it, as there are no machines like this in the world.”

Richard has met the Canadian scientist who made a robotic arm for the international space station. The arm, which is so famous it features on the Canadian $5 banknote, has a gripper on each end and can move around outside the space station.

In December this year the robot has to reach a milestone, with the machine working completely in a forest demonstration.

“We are looking for a commercial partner, such as a machinery company keen to take it to a bigger scale machine.”

He is hoping eventually they will develop a whole family of tree to tree robots to perform productive tasks in the forest.

“Large robots to fell trees, smaller robots to prune, monitor and measure trees, and perhaps even refuelling robots.

“They will still need humans to oversee their operation and maintenance, but they will be at a safe distance away from hazardous operations.”

Stefanie says the machine has potential for multiple functionality, apart from just cutting down trees.

She expects it might take five to eight years to commercialise the robot. Many other countries are interested in the idea too she says.