Javascript must be enabled for the correct page display

Lifelong 3D Object Recognition and Grasp Synthesis using Dual Memory Recurrent Self-Organization Networks

Santhakumar, Krishnakumar (2021) Lifelong 3D Object Recognition and Grasp Synthesis using Dual Memory Recurrent Self-Organization Networks. Master's Thesis / Essay, Artificial Intelligence.


Download (9MB) | Preview
[img] Text
Restricted to Registered users only

Download (129kB)


Humans learn to recognize and manipulate new objects in lifelong settings without forgetting the previously gained knowledge under non-stationary and sequential conditions. In this thesis, we proposed a hybrid model architecture consists of a dynamically growing dual-memory recurrent neural network (GDM) and an autoencoder to tackle both object recognition and grasping simultaneously. The GDM part is designed to recognize the object in both instances and categories levels, and the autoencoder network is responsible to extract a compact representation for a given object, which serves as input for the GDM learning, and is responsible to predict pixel-wise antipodal grasp configurations. We address the problem of catastrophic forgetting using the intrinsic memory replay, where the episodic memory periodically replays the neural activation trajectories in the absence of external sensory information. We generate a synthetic dataset to evaluate the proposed model since there is no standard sequential point cloud dataset that can be used in incremental learning scenarios exists. Experiment results demonstrated that the proposed model can learn both object representation and grasping simultaneously in continual learning scenarios.

Item Type: Thesis (Master's Thesis / Essay)
Supervisor name: Mohades Kasaei, S.H. and Wiering, M.A.
Degree programme: Artificial Intelligence
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 06 Sep 2021 07:13
Last Modified: 06 Sep 2021 07:13

Actions (login required)

View Item View Item