Resource-efficient lifelong robot learning

Equipping robots with the ability to learn a growing set of tasks over their operational lifetime—rather than focusing on mastering individual tasks—presents a significant challenge in robot learning. Lifelong learning robots often struggle with catastrophic forgetting when learning from changing input distributions, which causes the robot to forget old knowledge when learning new tasks. They are also expected to leverage previous knowledge to accelerate learning of new tasks without requiring complete retraining. This challenge is referred to as the stability-plasticity dilemma, where stability denotes the retention of old knowledge and plasticity refers to the acquisition of new knowledge.

Recent advancements in lifelong and continual learning have proposed three primary strategies to address this dilemma: 

• regularisation
• dynamic growth
• experience replay

However, these methods typically demand high storage and computational resources, leading to increased energy consumption for data storage, processing, and transmission. 

Robots also face limitations in onboard resources, making it difficult to support lifelong learning outside controlled environments and to retain and integrate experiences from diverse tasks and settings. Although some recent approaches show promise in improving efficiency in continual robot learning, they often do so at the cost of reduced performance compared to single-task models, where each task is learned with a dedicated model.

You will develop a continual on-device robot learning system that improves the trade-off between stability and plasticity while enhancing resource efficiency without compromising performance. The system targets deployment on resource-constrained, non-networked robotic platforms and contributes to sustainability by reducing carbon emissions through optimized operational efficiency.

This project is ideal if you are interested in sustainable AI, robotics, machine learning, and lifelong learning.