Abstract:
In recent years, the need for technological advancement has been increasing because of the declining birthrate and aging populations. Motion copying is one such technical...Show MoreMetadata
Abstract:
In recent years, the need for technological advancement has been increasing because of the declining birthrate and aging populations. Motion copying is one such technical development that can effectively reproduce human motion. However, the challenges of conventional motion-copying systems include the lack of operability and loss of touch sensation at the abstraction phase. Then, we focused on the force estimation by surface-EMG for abstracting force information. In this paper, we address this challenge by developing an explainable Artificial Intelligence (XAI) approach that can estimate force using surface-EMG and uses element description method to interpret and identify the components contributing towards that estimation. We evaluated the proposed XAI method across several real-world experiments that confirm its value and contribution towards innovations in motion-copying system.
Date of Conference: 19-21 June 2023
Date Added to IEEE Xplore: 31 August 2023
ISBN Information: