journal article·2023-01-01

Co-optimization of robotic design and skill inspired by human hand evolution

Bangchu Yang, Li Jiang, Guanjun Bao, Haoyong Yu, Xuanyi Zhou

Bioinspiration & Biomimetics· Volume 18· Issue 1
Publisher page

Abstract

Abstract During evolution of the human hand, evolutionary morphology has been closely related to behavior in complicated environments. Numerous researchers have revealed that learned skills have affected hand evolution. Inspired by this phenomenon, a co-optimization approach for underactuated hands is proposed that takes grasping skills and structural parameters into consideration. In our proposal, hand design, especially the underactuated mechanism, can be parameterized and shared with all the local agents. These mechanical parameters can be updated globally by the independent agents. In addition, we also train human-like ‘feeling’ of grasping: grasping stability is estimated in advance before the object drops, which can speed up grasping training. In this paper, our method is instantiated to address the optimization problem for the torsion spring mechanical parameters of an underactuated robotic hand with multi-actuators, and then the optimized results are transferred to the actual physical robotic hand to test the improvement of grasping. This collaborative evolution process leverages the dexterity of the multi-actuators and the adaptivity of the underactuated mechanism.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Robot Manipulation and Learning · Reinforcement Learning in Robotics · Motor Control and Adaptation

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (39)

Show source-provided references
1. Evolution of the human hand: approaches to acquiring, analysing and interpreting the anatomical evidence10.1046/j.1469-7580.2000.19710121.x ↗
2. Stone tools, language and the brain in human evolution10.1098/rstb.2011.0099 ↗
3. Estimating thumb–index finger precision grip and manipulation potential in extant and fossil primates10.1098/rsif.2015.0176 ↗
4. The impact of hand proportions on tool grip abilities in humans, great apes and fossil hominins: a biomechanical analysis using musculoskeletal simulation10.1016/j.jhevol.2018.10.001 ↗
5. Assessing the manipulative potentials of monkeys, apes and humans from hand proportions: implications for hand evolution10.1098/rspb.2016.1923 ↗
6. The evolution of human and ape hand proportions10.1038/ncomms8717 ↗
7. The human extensor digitorum profundus muscle with comments on the evolution of the primate hand10.1007/bf02382073 ↗
8. Exploring the existence of better hands for manipulation than the human hand based on hand proportions10.1016/j.jtbi.2017.12.026 ↗
9. Morphology and behaviour: functional links in development and evolution10.1098/rstb.2011.0035 ↗
10. Data-driven optimization for underactuated robotic hands
11. Shadow delivers first hand10.1108/01439910510573237 ↗
12. Learning dexterous in-hand manipulation10.1177/0278364919887447 ↗
13. The Hannes hand prosthesis replicates the key biological properties of the human hand10.1126/scirobotics.abb0467 ↗
14. Multisensory five-finger dexterous hand: the DLR/HIT hand II
15. Integrated linkage-driven dexterous anthropomorphic robotic hand10.1038/s41467-021-27261-0 ↗
16. Toward dexterous manipulation with augmented adaptive synergies: the Pisa/IIT SoftHand 210.1109/tro.2018.2830407 ↗
17. Design principle of a dual-actuated robotic hand with anthropomorphic self-adaptive grasping and dexterous manipulation abilities10.1109/TRO.2021.3132532 ↗
18. A modular multisensory prosthetic hand
19. Force analysis of whole hand grasp by multifingered robotic hand
21. Design and testing of a selectively compliant underactuated hand10.1177/0278364913518997 ↗
22. A hand-centric classification of human and robot dexterous manipulation10.1109/TOH.2012.53 ↗
23. Grasp frequency and usage in daily household and machine shop tasks10.1109/toh.2013.6 ↗
24. The grasp taxonomy of human grasp types10.1109/THMS.2015.2470657 ↗
25. Modelling natural and artificial hands with synergies10.1098/rstb.2011.0152 ↗
26. Geometric design optimization of an under-actuated tendon-driven robotic gripper10.1016/j.rcim.2017.09.012 ↗
27. Experimental evaluation of the Schunk 5-finger gripping hand for grasping tasks10.1109/ROBIO.2014.7090710 ↗
28. Reinforcement learning for improving agent design10.1162/artl_a_00301 ↗
29. Jointly learning to construct and control agents using deep reinforcement learning
30. Data-efficient co-adaptation of morphology and behaviour with deep reinforcement learning
31. Embodied intelligence via learning and evolution10.1038/s41467-021-25874-z ↗
32. Living object grasping using two-stage graph reinforcement learning10.1109/lra.2021.3060636 ↗
33. The complexity of Markov decision processes10.1287/moor.12.3.441 ↗
34. Dexterous manipulation for multi-fingered robotic hands with reinforcement learning: a review10.3389/fnbot.2022.861825 ↗
35. Assessing grasp stability based on learning and haptic data10.1109/tro.2011.2132870 ↗
36. Soft actor-critic: off-policy maximum entropy deep reinforcement learning with a stochastic actor
37. Design and implementation of an anthropomorphic hand for replicating human grasping functions10.1109/tro.2016.2558193 ↗
38. A comprehensive grasp taxonomy
39. The YCB object and model set: towards common benchmarks for manipulation research