Robotics at Robinson
Our robotics team is pioneering new frontiers in autonomous mobile robotics.
Expanding applications beyond structured environments
To date, autonomous mobile robots have primarily been deployed in structured environments such as warehouses, logistics hubs, and controlled outdoor areas like electricity substations, mining sites, precision agriculture, and lawn-mowing services.
While these robots have proven effective in such settings, there is a growing demand for robots that can function in far more unpredictable and dynamic environments.
Industries such as agriculture, forestry, disaster response, underwater exploration, and planetary missions present environments that are not only unstructured but also highly variable and potentially hazardous. Robust mobile robots capable of functioning effectively in these challenging settings could transform these sectors.
Overcoming challenges in unstructured environments
Our research addresses the need for autonomous robots that can operate in more challenging outdoor environments where traditional robotics solutions fall short.
These environments often include rugged terrain, areas with unreliable or unavailable GNSS signals for localisation, and adverse weather conditions. Additionally, these settings lack prior information about the environment, and robots must operate without constraints on their position relative to objects and people. Additional hazards exist; for instance, in forestry, slipping, tripping, and falling are the second-largest causes of reported lost-time injuries due to hazardous underfoot conditions—risks that are equally applicable to robots.
Our team is developing innovative control algorithms and robust sensing systems to tackle these challenges. These solutions are designed to ensure that robots maintain safety, stability, and high efficiency, even in the most demanding conditions.
Our focus on smart sensing, perception, and control allows us to develop real-world solutions that do not rely solely on physical design but on enhancing the robot's ability to sense and adapt to its environment.
Innovative solutions for stability and navigation
A key aspect of our research is enhancing autonomous robots' stability and navigation capabilities, particularly in unstructured environments.
Uneven and slippery terrain can cause robots to lose balance, tip-over, or deviate from their intended path. To mitigate these risks, some of our recent projects have focused on terra-mechanics and wheel-terrain interaction to address robot slip, skid, and tip-over, using sensors unaffected by weather conditions.
We are also advancing algorithms for motion control and path planning and integrating AI-driven perception systems that allow robots to maintain precise localisation and adapt to sudden changes, even in GNSS-denied areas like dense forests and urban canyons.
Enhancing sensor reliability and machine learning capabilities
Some common sensor types often fail or provide inaccurate data under non-ideal weather conditions, such as rain or fog, posing significant challenges for autonomous operation.
Our research focuses on developing solutions using robust, cost-effective sensory inputs that are less susceptible to environmental interference. By integrating artificial intelligence, including machine learning and deep learning models, we enable robots to predict and respond to obstacles in real-time, creating an artificial "mind" that allows them to perceive, model, and autonomously navigate complex environments. This significantly reduces the need for human intervention, making robots more autonomous and efficient.
Transformative applications across multiple industries
The potential applications of our research are vast and transformative.
In agriculture and forestry, autonomous robots could navigate and perform tasks in complex terrains, reducing the need for human workers in dangerous conditions and increasing efficiency.
In disaster response and search and rescue operations, robots could enhance safety by operating in hazardous zones where human entry is too risky.
Our research can also be readily adapted for planetary exploration and underwater robotics, where robots must navigate unknown, rugged terrains.
Future research could expand into terrain characterisation and traversability analysis, further enhancing the robots' ability to navigate complex environments.
A unique New Zealand advantage—research in natural laboratories
One of the unique advantages of conducting robotics research in New Zealand is our unparalleled access to some of the world's most diverse and challenging outdoor environments. From ocean floors to snowcapped mountains and urban jungles to rainforests, our researchers can test and refine their robotic systems in real-world conditions close to the lab. This proximity allows us to push the boundaries of what our robots can achieve, ensuring they are equipped to handle a wide range of environments globally.