Autonomous detection and sorting of litter using deep learning and soft robotic grippers
Road infrastructure is one of the most vital assets of any country. Keeping the road infrastructure clean and unpolluted is important for ensuring road safety and reducing environmental risk. However, roadside litter picking is an extremely laborious, expensive, monotonous and hazardous task. Automa...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-12-01
|
| Series: | Frontiers in Robotics and AI |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/frobt.2022.1064853/full |
| _version_ | 1811213502534123520 |
|---|---|
| author | Elijah Almanzor Nzebo Richard Anvo Nzebo Richard Anvo Thomas George Thuruthel Fumiya Iida |
| author_facet | Elijah Almanzor Nzebo Richard Anvo Nzebo Richard Anvo Thomas George Thuruthel Fumiya Iida |
| author_sort | Elijah Almanzor |
| collection | DOAJ |
| description | Road infrastructure is one of the most vital assets of any country. Keeping the road infrastructure clean and unpolluted is important for ensuring road safety and reducing environmental risk. However, roadside litter picking is an extremely laborious, expensive, monotonous and hazardous task. Automating the process would save taxpayers money and reduce the risk for road users and the maintenance crew. This work presents LitterBot, an autonomous robotic system capable of detecting, localizing and classifying common roadside litter. We use a learning-based object detection and segmentation algorithm trained on the TACO dataset for identifying and classifying garbage. We develop a robust modular manipulation framework by using soft robotic grippers and a real-time visual-servoing strategy. This enables the manipulator to pick up objects of variable sizes and shapes even in dynamic environments. The robot achieves greater than 80% classified picking and binning success rates for all experiments; which was validated on a wide variety of test litter objects in static single and cluttered configurations and with dynamically moving test objects. Our results showcase how a deep model trained on an online dataset can be deployed in real-world applications with high accuracy by the appropriate design of a control framework around it. |
| first_indexed | 2024-04-12T05:47:46Z |
| format | Article |
| id | doaj.art-7b71d416213d4853ac7a56c9101dd957 |
| institution | Directory Open Access Journal |
| issn | 2296-9144 |
| language | English |
| last_indexed | 2024-04-12T05:47:46Z |
| publishDate | 2022-12-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Robotics and AI |
| spelling | doaj.art-7b71d416213d4853ac7a56c9101dd9572022-12-22T03:45:25ZengFrontiers Media S.A.Frontiers in Robotics and AI2296-91442022-12-01910.3389/frobt.2022.10648531064853Autonomous detection and sorting of litter using deep learning and soft robotic grippersElijah Almanzor0Nzebo Richard Anvo1Nzebo Richard Anvo2Thomas George Thuruthel3Fumiya Iida4The Bio-Inspired Robotics Lab, Department of Engineering, University of Cambridge, Cambridge, United KingdomThe Bio-Inspired Robotics Lab, Department of Engineering, University of Cambridge, Cambridge, United KingdomCostain Group PLC, Cambridge, United KingdomThe Bio-Inspired Robotics Lab, Department of Engineering, University of Cambridge, Cambridge, United KingdomThe Bio-Inspired Robotics Lab, Department of Engineering, University of Cambridge, Cambridge, United KingdomRoad infrastructure is one of the most vital assets of any country. Keeping the road infrastructure clean and unpolluted is important for ensuring road safety and reducing environmental risk. However, roadside litter picking is an extremely laborious, expensive, monotonous and hazardous task. Automating the process would save taxpayers money and reduce the risk for road users and the maintenance crew. This work presents LitterBot, an autonomous robotic system capable of detecting, localizing and classifying common roadside litter. We use a learning-based object detection and segmentation algorithm trained on the TACO dataset for identifying and classifying garbage. We develop a robust modular manipulation framework by using soft robotic grippers and a real-time visual-servoing strategy. This enables the manipulator to pick up objects of variable sizes and shapes even in dynamic environments. The robot achieves greater than 80% classified picking and binning success rates for all experiments; which was validated on a wide variety of test litter objects in static single and cluttered configurations and with dynamically moving test objects. Our results showcase how a deep model trained on an online dataset can be deployed in real-world applications with high accuracy by the appropriate design of a control framework around it.https://www.frontiersin.org/articles/10.3389/frobt.2022.1064853/fullAI-driven controlsoft roboticsdeep learningvisual servoinglitter picking |
| spellingShingle | Elijah Almanzor Nzebo Richard Anvo Nzebo Richard Anvo Thomas George Thuruthel Fumiya Iida Autonomous detection and sorting of litter using deep learning and soft robotic grippers Frontiers in Robotics and AI AI-driven control soft robotics deep learning visual servoing litter picking |
| title | Autonomous detection and sorting of litter using deep learning and soft robotic grippers |
| title_full | Autonomous detection and sorting of litter using deep learning and soft robotic grippers |
| title_fullStr | Autonomous detection and sorting of litter using deep learning and soft robotic grippers |
| title_full_unstemmed | Autonomous detection and sorting of litter using deep learning and soft robotic grippers |
| title_short | Autonomous detection and sorting of litter using deep learning and soft robotic grippers |
| title_sort | autonomous detection and sorting of litter using deep learning and soft robotic grippers |
| topic | AI-driven control soft robotics deep learning visual servoing litter picking |
| url | https://www.frontiersin.org/articles/10.3389/frobt.2022.1064853/full |
| work_keys_str_mv | AT elijahalmanzor autonomousdetectionandsortingoflitterusingdeeplearningandsoftroboticgrippers AT nzeborichardanvo autonomousdetectionandsortingoflitterusingdeeplearningandsoftroboticgrippers AT nzeborichardanvo autonomousdetectionandsortingoflitterusingdeeplearningandsoftroboticgrippers AT thomasgeorgethuruthel autonomousdetectionandsortingoflitterusingdeeplearningandsoftroboticgrippers AT fumiyaiida autonomousdetectionandsortingoflitterusingdeeplearningandsoftroboticgrippers |