Domain Adaptation for Outdoor Robot Traversability Estimation from RGB data with Safety-Preserving Loss

Palazzo, Simone; Guastella, Dario Calogero; Cantelli, Luciano; Spadaro, Paolo; Rundo, Francesco; Muscato, Giovanni; Giordano, Daniela; Spampinato, Concetto

doi:10.1109/iros45743.2020.9341044

Cited by 15 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar to Wulfmeier et al (2017), the method proposed in Palazzo et al (2020) aims to enhance outdoor robot traversability through domain adaptation using gradient reversal technique. Their method adds a domain classification mod-ule alongside the traversability prediction module and reverses the gradients from domain classifier loss when propagating through the image encoder module.…”

Section: B Offline Methodsmentioning

confidence: 99%

Learning for Robot Decision Making under Distribution Shift: A Survey

Paudel¹

2022

Preprint

View full text Add to dashboard Cite

With the recent advances in the field of deep learning, learning-based methods are widely being implemented in various robotic systems that help robots understand their environment and make informed decisions to achieve a wide variety of tasks or goals. However, learning-based methods have repeatedly been shown to have poor generalization when they are presented with inputs that are different from those during training leading to the problem of distribution shift. Any robotic system that employs learning-based methods is prone to distribution shift which might lead the agents to make decisions that lead to degraded performance or even catastrophic failure. In this paper, we discuss various techniques that have been proposed in the literature to aid or improve decision making under distribution shift for robotic systems. We present a taxonomy of existing literature and present a survey of existing approaches in the area based on this taxonomy. Finally, we also identify a few open problems in the area that could serve as future directions for research.

show abstract

Section: B Offline Methodsmentioning

confidence: 99%

Learning for Robot Decision Making under Distribution Shift: A Survey

Paudel¹

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The path included a 10m steep slope (50 o ) followed by a 5-m flat area, which transitioned into a 30-m variable-slope region (20-50 o ) to the top. The path also included some minor (about 10•) cross-slopes [15], [57], [58], [59]. The terrain was hard soil covered with light vegetation [15].…”

Section: Testing Methodsmentioning

confidence: 99%

A Review of Mobile Robot Navigation System for Volcano Monitoring Application

Evita¹

2021

ijp

View full text Add to dashboard Cite

Volcano is a geological environment including magma, eruption, volcanic edifice and its basements. For continuous monitoring after eruption, a mobile robot could be proposed as an alternative to prevent hazardous effect to volcanologist who perform up close monitoring. In this paper, the robots were divided into 3 types according to their different structures: legged, track-legged and wheeled mobile robots. Meanwhile, the navigation system were implemented in 4 steps suitable for volcano condition: environment mapping, trajectory design, motion control and obstacle avoidance. These navigation system also tested in different locations: indoor, outdoor and real volcano with different testing method for these robots. The testing result was discussed in robot kinematics parameter such as trajectory, velocity, slope angle, rollover and sideslip angels.

show abstract

“…Then, VUNet predicts the traversability of the image by applying the synthesized future images to GONet [12]. In [14], the traversability of the area surrounding a mobile robot was estimated by dividing the image into several bins and training the deep learning network to learn the reachable area from each bin. This paper demonstrates good performance when it is applied to mountains, plains, or at least on country roads.…”

Section: Related Work a Traversable Region Identification Using An Rg...mentioning

confidence: 99%

Street Floor Segmentation for a Wheeled Mobile Robot

2022

View full text Add to dashboard Cite

In urban cities, the information about the type or class of street floors enables a wheeled mobile robot to perform many tasks ranging from traversability region identification, localization and the choice of wheel control strategy. In this paper, we considered a new task named as street floor segmentation (SFS) using an RGB camera. The SFS can be considered as the generalized problem of the existing traversability region identification problem in urban situations. Our SFS has two special classes for the possible application to the traversability region identification and they are traversable and non-traversable curbs. The SFS using an RGB camera is implemented using a real-time semantic segmentation (SS) network. A booster module named as Dynamic Context-based Refinement Module (DCRM) was developed to enhance the performance of the SFS. Our network was applied to real-world applications, and its validity is demonstrated through experiment.

show abstract

Domain Adaptation for Outdoor Robot Traversability Estimation from RGB data with Safety-Preserving Loss

Cited by 15 publications

References 15 publications

Learning for Robot Decision Making under Distribution Shift: A Survey

Learning for Robot Decision Making under Distribution Shift: A Survey

A Review of Mobile Robot Navigation System for Volcano Monitoring Application

Street Floor Segmentation for a Wheeled Mobile Robot

Contact Info

Product

Resources

About