Adaptive learning robots to complement the human workforce

Image of computer configuration of robotic tasks in unknown environments.

Imagine a world where robots can be trained to work alongside humans in factories and the outdoors in a safe and flexible way. This Spearhead project is developing software and prototypes to make this happen.

Snapshot

  • Industry is increasingly demanding robots that can safely complete tasks and adapt to their environment as they work in close quarters with humans.
  • Guided by industry advisors, our researchers are developing algorithms, software and robotic prototypes to build smarter robots with safety front of mind.
  • This project represents an invaluable opportunity to build New Zealand’s capability in robotics, while generating new automation options in everything from horticulture to high-value manufacturing.

 

An example of easily adaptive configuration and hardware features needed for changing robotic tasks.

Helping humans and robots work together

Robots are already present in industrial settings such as car manufacturing, automated milking, and meat processing. Newer forms of robots are now needed to support other industries including small-medium primary industries, such as seafood, horticultural and forestry sectors. Advances in robotic technology will lead to higher workplace safety, and different types of work for humans and robots.

 

Real-time volumetric models for integrating robotic perception and other data required for the robot to ‘perceive’ the world and conduct flexible tasks.

Adaptive learning in robots - how does it work?

To develop this technology, our researchers will investigate how sensors and artificial intelligence can allow robots to perceive and understand their surroundings, flexibly handle new situations through learning or training by humans or other robots, and work in challenging environments.

This involves developing new algorithms and software to analyse sensor data coming from robots and their environment, to inform their work. The robots will feature ‘adaptive perception’, which gives them the ability to learn when and how to modify their activity as they complete their tasks.

A user interface based on 3D/4D mapping will be developed for the Robot Operating System (ROS) the team is using for its robotic prototypes. Hardware models will be created to show how a robot’s physical characteristics can morph ‘on the fly’ as required.

Safety will be a particular focus with systems created to allow robots to take safety measures when working at high speed in human workplaces.

Developing a niche market

It is envisaged that after 8-10 years, New Zealand will have developed niche adaptable robots; in contrast to mass-produced single-task robots. The technology is expected to help the country’s industries thrive globally and create an international hub for innovative robotics development.

Introducing robots to unstructured environments and to closer working interaction with humans poses unique challenges. They require new innovations to tackle, but with some of the country’s leading researchers committed to the task, this project is well placed to help New Zealand prepare for a future where humans and robots work together in harmony.

The team 

  • Spearhead Leader: Dr Armin Werner – Lincoln Agritech, Group Manager Precision Agriculture
  • Associate Professor Will Browne – Victoria University Wellington
  • Associate Professor Johan Potgieter – Massey University
  • The project involves researchers from Lincoln Agritech and SCION, as well as Auckland, Victoria, Massey, Canterbury and Otago Universities.

Top image: Computer configuration of robotic tasks in unknown environments to identify future options in handling robotic tasks in unknown environments.