Deep Robot Path Planning from Demonstrations for Breast Cancer Examination

Crivellari, Marta; Sanni, Oluwatoyin; Zanchettin, Andrea Maria; Esfahani, Amir Ghalamzan

doi:10.1007/978-3-030-89177-0_27

Cited by 2 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…End-to-end learning such as Inverse Reinforcement Learning (Levine, Popovic, andKoltun 2011), vision-based Model Predictive Control (Finn andLevine 2017), or Behaviour cloning (Rahmatizadeh et al 2018) usually suffer from lack of generalisation and they are only applied to a class of tasks involved complicated motion control but not complex motion planning. (Crivellari et al 2021) presents deep-LfD models for the ARBP with a simple breast model and without visual inputs. End-to-end LfD enables generating control commands directly from raw image data, e.g., (Rahmatizadeh et al 2018) proposed behaviour cloning probabilistic generative model.…”

Section: Related Workmentioning

confidence: 99%

Deep Movement Primitives: Toward Breast Cancer Examination Robot

Sanni

Bonvicini

Khan

et al. 2022

AAAI

Self Cite

View full text Add to dashboard Cite

Breast cancer is the most common type of cancer worldwide. A robotic system performing autonomous breast palpation can make a significant impact on the related health sector worldwide. However, robot programming for breast palpating with different geometries is very complex and unsolved. Robot learning from demonstrations (LfD) reduces the programming time and cost. However, the available LfD are lacking the modelling of the manipulation path/trajectory as an explicit function of the visual sensory information. This paper presents a novel approach to manipulation path/trajectory planning called deep Movement Primitives that successfully generates the movements of a manipulator to reach a breast phantom and perform the palpation. We show the effectiveness of our approach by a series of real-robot experiments of reaching and palpating a breast phantom. The experimental results indicate our approach outperforms the state-of-the-art method.

show abstract