Year: 2024
Type: article
Abstract: Predicting long-term visual outcomes for robot manipulation tasks is crucial for enabling robots to anticipate future changes in their environment and plan optimal actions accordingly. This research is presents a novel approach to long-term visual prediction using vision-based techniques and deep learning models. We propose a hybrid convolutional neural network (CNN) and recurrent neural network (RNN) architecture that combines spatial feature extraction with temporal modeling to predict future visual states accurately. The predictive model is trained on annotated datasets of robot manipulation sequences, allowing it to learn complex spatial and temporal relationships in the data. Experimental results demonstrate the effectiveness of the proposed approach in accurately predicting long-term visual outcomes for a variety of manipulation tasks.
Authors: Munir Ahmad, Taib Ali, Nabeel Ali Khan, Asima Afzal, Talha Bin Sohail, Tehmina Shahid
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4408851219