perform mechanism identification

This study addresses the challenge of accurately identifying dynamic models for low-cost robotic arms, which often suffer from parameter redundancy and physical infeasibility. Focusing on the CRANE-X7 manipulator, the work proposes a reproducible, physically consistent parameter identification pipeline. It begins with a simplified rigid-body model and a structured excitation trajectory, then integrates inverse dynamics regression, ordinary least squares (OLS), semidefinite programming (SDP) feasibility projection, and closed-loop input error (CLIE) optimization. A novel two-stage filtering mechanism is introduced—combining statistical centrality-based selection with full-pose positive definiteness auditing—to ensure both statistical consistency and physical plausibility. The estimated parameters progressively converge from OLS through SDP to CLIE, achieving high prediction accuracy while rigorously adhering to physical constraints.

Planar four-bar linkage dimensional synthesis is a classic inverse kinematics problem requiring the inference of mechanism parameters from prescribed trajectory constraints. This paper proposes a data-driven, end-to-end supervised learning framework that integrates Long Short-Term Memory (LSTM) networks with a type-aware Mixture-of-Experts (MoE) architecture, enabling unified modeling and generation across multiple linkage types—including crank-rocker, double-crank, and double-rocker mechanisms. We introduce a novel motion-simulation-based metric and combine synthetic-data training with kinematic feedback to refine generation quality. Experiments demonstrate that our method efficiently produces high-accuracy linkage parameters free from assembly or mobility defects. It significantly lowers the design barrier for non-expert users while improving both synthesis efficiency and practical feasibility.

This work addresses the challenge of safe autonomous manipulation in unknown environments where prior knowledge of constraint types and stiffness is unavailable. The authors propose an online path-planning algorithm that simultaneously identifies constraint types, estimates their screw parameters, and constructs a global stiffness model by fusing force and pose feedback in real time during exploration. By integrating stiffness tensor analysis, eigenvector mapping, and a library of canonical constraint patterns—such as hinges and planar contacts—the method achieves, for the first time, real-time identification and characterization of common mechanical constraints without any prior information. Simulations and physical experiments demonstrate that the system accurately recognizes compliant constraints like flexible hinges, enabling safe human–robot collaboration in applications such as surgical organ retraction.

研究通过基于浮动参考框架的柔性多体建模方法，解决了过约束空间连杆在工业应用中的装配精度问题，并验证了其自组装倾向。

This work addresses the challenge of high-precision joint estimation of robotic states and physical parameters under constraints and disturbances by proposing a Bayesian system identification framework. The approach embeds physically consistent inverse dynamics, contact and closed-loop kinematic constraints, and a comprehensive joint friction model as hard equality constraints within the estimation process. Energy-based regressors are incorporated to enhance parameter observability, while prior constraints on inertial and actuation parameters are explicitly supported. A constrained Riccati recursion algorithm preserving banded matrix structure is devised to achieve linear time complexity for efficient computation. Evaluations in simulation and experiments on the Unitree B1 robot demonstrate that, compared to forward-dynamics and decoupled estimation methods, the proposed framework significantly accelerates convergence, reduces errors in inertial and friction parameter estimates, improves contact consistency, and substantially enhances trajectory tracking performance in complex terrains when integrated into model predictive control (MPC).

本文扩展了刚体动力学一阶导数算法，以适应闭链系统，并通过约束嵌入方法解决了现有算法在处理复杂关节类型时的局限性。

This study addresses the challenge of dynamic modeling for mechanisms with variable topology, particularly when constraints such as joint locking, static friction, or ideal contact lead to abrupt changes in degrees of freedom. To ensure physically consistent and continuous dynamic behavior during topological transitions, the work proposes a set of physically coherent switching conditions. Building upon this foundation, two nonsmooth dynamics frameworks are developed: one based on redundant coordinates employing projected equations of motion, and another using minimal coordinates formulated via Voronets equations. The computational characteristics of both approaches are systematically compared. The proposed methodology is successfully validated on a planar 3R mechanism and a 6-DOF industrial manipulator under joint-locking scenarios, significantly enhancing the accuracy and feasibility of forward dynamics simulations for complex systems with varying topology.

This study addresses the challenge of accurately modeling the dynamics of flexible two-degree-of-freedom robotic arms, where rigid-body assumptions fail to capture link flexibility and unmodeled residual dynamics. To overcome this limitation, the authors propose a semi-parametric hybrid dynamical modeling framework that augments rigid-body dynamics with a Gaussian Mixture Model (GMM) to learn residual terms, while employing a purely data-driven kinematic regression as a baseline. This approach synergistically integrates physical priors with data-driven mechanisms, transcending the constraints of conventional fully parametric models in flexible systems. Experimental results on an open-source dataset demonstrate that the data-driven component—combined with regularization and least-squares estimation—significantly enhances torque prediction accuracy, thereby validating the efficacy and superiority of the proposed hybrid model.

This work addresses the trade-off between infinite-dimensional modeling, compact state representation, and uncertainty quantification in reconstructing the shape of continuum robots from sparse, noisy sensor data. The authors propose a low-dimensional state estimation approach based on factor graphs, leveraging geometrically varying strain (GVS) to parameterize the strain field. For the first time, closed-form kinematic constraints derived via the Magnus expansion are incorporated as factors into the graphical model, establishing a geometric prior—grounded in Cosserat rod theory—between strain and pose. This enables compact, probabilistic, and modular state inference. In simulations on a 0.4-meter tendon-driven continuum robot, the method achieves average positional errors below 2 mm across three sensing configurations; when using only position measurements, it reduces orientation error by a factor of six compared to a Gaussian process regression baseline.

本文提出一种新方法，通过逆函数定理计算解析IK参数化的梯度，以解决机器人在运动学约束下的轨迹规划问题。