Hybrid Adaptive Control Strategy for Continuum Surgical Robot Under External Load

Wang, Ziwen; Teng, W. L.; Zhao, Baoliang; He, Yucheng; Hu, Ying; Li, Bing; Zhang, Peng; Meng, Max Q.-H.

doi:10.1109/lra.2021.3057558

Cited by 41 publications

(19 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When estimating the error, it is necessary to consider that the authors provide the error statistics regardless of the robot workspace. However, the described method suggested finding the best robot configuration under defined conditions with closed-form expressions, avoiding neural networks [41] or robot motion learning [42] techniques. However, as this work clarifies, the accuracy of the end-effector's state is the main trade-off.…”

Section: Discussionmentioning

confidence: 99%

“…The estimated second segment's endpoint and its endpoint's orientation vector exist in the first segment coordinate system (Figure 12). Translation of the second robot segment endpoint , arc center point , and orientation vector back to the second segment coordination system requires their rotation according to the first segment counter-clockwise quaternion rotation matrix in Equation (41). After that, , and add to the first segment endpoint .…”

Section: Calculation Of Arc Segments' Endpoints and Orientationsmentioning

confidence: 99%

See 1 more Smart Citation

Synchronized Motion Profiles for Inverse-Dynamics-Based Online Control of Three Inextensible Segments of Trunk-Type Robot Actuators

et al. 2021

View full text Add to dashboard Cite

This study proposes a novel method for the positioning and spatial orientation control of three inextensible segments of trunk-type robots. The suggested algorithm imposes a soft constraint assumption for the end-effector’s endpoint and a mandatory constraint on its direction. Simultaneously, the algorithm by-design enforces nonholonomic features on the robot segments in the form of arcs. An approximate robot spine curve is the key to the final robot state configuration based on the given conditions. The numeric simulation showed acceptable (less than 1 s) performance for single-core processing tasks. The parametric method finds the best proximate robot state solution and represents the gray box model in addition to existing learning or black-box inverse dynamics approaches. This study also shows that a multiple inverse kinematics answer constructs a single inverse dynamics solution that defines the robot actuators’ motion profiles, synchronized in time. Finally, this text presents rotational expressions and their outlines for controlling the manipulator’s tendons.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Calculation Of Arc Segments' Endpoints and Orientationsmentioning

confidence: 99%

Synchronized Motion Profiles for Inverse-Dynamics-Based Online Control of Three Inextensible Segments of Trunk-Type Robot Actuators

et al. 2021

View full text Add to dashboard Cite

show abstract

“…After reaching a target location, the robot would move back to its home (straight) configuration. Three different target points [(0,8) mm,(0,16) mm, (10,16) mm] were selected to ensure the robot was in contact and pressed the foam with varying degree of forces. Each point was tracked 20 times.…”

Section: Trajectory Tracking With and Without Shieldingmentioning

confidence: 99%

“…Hybrid offline and online learning methods have also been proposed. The offline model could be trained using conventional model, finite element analysis, and experimental data on neural networks while online adaptive scheme is developed to update the parameters of locally weighted projection regression, local Gaussian process regression models [9] or proportional-integral-derivative (PID) controller [10]. While online learning methods adapt to environmental interactions, it requires training with a large data set.…”

Section: Introductionmentioning

confidence: 99%

Towards Safe Control of Continuum Manipulator Using Shielded Multiagent Reinforcement Learning

Yan

et al. 2021

Preprint

View full text Add to dashboard Cite

Continuum robotic manipulators are increasingly adopted in minimal invasive surgery. However, their nonlinear behavior is challenging to model accurately, especially when subject to external interaction, potentially leading to poor control performance. In this letter, we investigate the feasibility of adopting a model-free multiagent reinforcement learning (RL), namely multiagent deep Q network (MADQN), to control a 2-degree of freedom (DoF) cable-driven continuum surgical manipulator. The control of the robot is formulated as a one-DoF, one agent problem in the MADQN framework to improve the learning efficiency. Combined with a shielding scheme that enables dynamic variation of the action set boundary, MADQN leads to efficient and importantly safer control of the robot. Shielded MADQN enabled the robot to perform point and trajectory tracking with submillimeter root mean square errors under external loads, soft obstacles, and rigid collision, which are common interaction scenarios encountered by surgical manipulators. The controller was further proven to be effective in a miniature continuum robot with high structural nonlinearitiy, achieving trajectory tracking with submillimeter accuracy under external payload.

show abstract

“…Joint types for the cable-driven hyper-redundant robot mainly can be classified into the 1-DOF joint and 2-DOF joint [12]. e 1-DOF joints mainly include the revolute joint [13,14], the flexible beam [15,16], and the cylindrical rolling joint [17,18], while the multi-DOF joints mainly include the universal joint [19,20], the flexible backbone [21,22], and the spherical rolling joint. Based on the flexible backbone, Li et al developed a 2-DOF flexible endoscope driven by multiple cables, which is more dexterous than rigid endoscopes [23].…”

Section: Introductionmentioning

confidence: 99%

Design and Validation of a Novel Cable-Driven Hyper-Redundant Robot Based on Decoupled Joints

Huang

Liu

Yin

et al. 2021

Journal of Robotics

View full text Add to dashboard Cite

In most of the prior designs of conventional cable-driven hyper-redundant robots, the multiple degree-of-freedom (DOF) bending motion usually has bending coupling effects. It means that the rotation output of each DOF is controlled by multiple pairs of cable inputs. The bending coupling effect will increase the complexity of the driving mechanism and the risk of slack in the driving cables. To address these problems, a novel 2-DOF decoupled joint is proposed by adjusting the axes distribution of the universal joints. Based on the decoupled joint, a 4-DOF hyper-redundant robot with two segments is developed. The kinematic model of the robot is established, and the workspace is analyzed. To simplify the driving mechanism, a kinematic fitting approach is presented for both proximal and distal segments and the mapping between the actuator space and the joint space is significantly simplified. It also leads to the simplification of the driving mechanism and the control system. Furthermore, the cable-driven hyper-redundant robot prototype with multiple decoupled joints is established. The experiments on the robot prototype verify the advantages of the design.

show abstract

Hybrid Adaptive Control Strategy for Continuum Surgical Robot Under External Load

Cited by 41 publications

References 22 publications

Synchronized Motion Profiles for Inverse-Dynamics-Based Online Control of Three Inextensible Segments of Trunk-Type Robot Actuators

Synchronized Motion Profiles for Inverse-Dynamics-Based Online Control of Three Inextensible Segments of Trunk-Type Robot Actuators

Towards Safe Control of Continuum Manipulator Using Shielded Multiagent Reinforcement Learning

Design and Validation of a Novel Cable-Driven Hyper-Redundant Robot Based on Decoupled Joints

Contact Info

Product

Resources

About