Robot Dynamics with URDF &amp; CasADi

Johannessen, Lill Maria Gjerde; Arbo, Mathias Hauan; Gravdahl, Jan Tommy

doi:10.1109/iccma46720.2019.8988702

Cited by 11 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of this tool requires the recursive Newton-Euler inverse dynamics of the robot to be written in a symbolic form. For this reason, we adopt urdf2casadi, a software library for computing functions of the robot dynamics that can be adopted with symbolic expressions in the CasADi framework, based on a URDF description of the robot [44]. The parameters of the dynamic model of the Franka arm have been taken from the model that was experimentally verified in [38].…”

Section: T-time R-idle C-act-ws R-stopsmentioning

confidence: 99%

Enhancing fluency and productivity in human-robot collaboration through online scaling of dynamic safety zones

Scalera

Giusti

Vidoni

et al. 2022

Int J Adv Manuf Technol

View full text Add to dashboard Cite

Industrial collaborative robotics is promising for manufacturing activities where the presence of a robot alongside a human operator can improve operator’s working conditions, flexibility, and productivity. A collaborative robotic application has to guarantee not only safety of the human operator, but also fluency in the collaboration, as well as performance in terms of productivity and task time. In this paper, we present an approach to enhance fluency and productivity in human-robot collaboration through online scaling of dynamic safety zones. A supervisory controller runs online safety checks between bounding volumes enclosing robot and human to identify possible collision dangers. To optimize the sizes of safety zones enclosing the manipulator, the method minimizes the time of potential stop trajectories considering the robot dynamics and its torque constraints, and leverages the directed speed of the robot parts with respect to the human. Simulations and experimental tests on a seven-degree-of-freedom robotic arm verify the effectiveness of the proposed approach, and collaborative fluency metrics show the benefits of the method with respect to existing approaches.

show abstract

Section: T-time R-idle C-act-ws R-stopsmentioning

confidence: 99%

Enhancing fluency and productivity in human-robot collaboration through online scaling of dynamic safety zones

Scalera

Giusti

Vidoni

et al. 2022

Int J Adv Manuf Technol

View full text Add to dashboard Cite

show abstract

“…1) Optimization problem: Neglecting external forces, the dynamics of the system are defined by the equation of motion as τ = M (q)q + C(q, q)q + g(q), which is composed of the mass matrix M , the Coriolis matrix C and the gravitational forces g [6]. Various approaches to compute the forward dynamics have been proposed [10]. The forward dynamics can be discretized to obtain the transition function…”

Section: Appendix a Model Predictive Controllermentioning

confidence: 99%

“…The parameter setting is summarized in Table III The optimization problem is solved using the nonlinear solver proposed in [31] and the corresponding implementation [5]. The forward dynamics are computed using [10].…”

Section: Appendix a Model Predictive Controllermentioning

confidence: 99%

Adaptation through prediction: multisensory active inference torque control

Meo¹,

Franzese²,

Pezzato³

et al. 2021

Preprint

View full text Add to dashboard Cite

Adaptation to external and internal changes is major for robotic systems in uncertain environments. Here we present a novel multisensory active inference torque controller for industrial arms that shows how prediction can be used to resolve adaptation. Our controller, inspired by the predictive brain hypothesis, improves the capabilities of current active inference approaches by incorporating learning and multimodal integration of low and high-dimensional sensor inputs (e.g., raw images) while simplifying the architecture. We performed a systematic evaluation of our model on a 7DoF Franka Emika Panda robot arm by comparing its behavior with previous active inference baselines and classic controllers, analyzing both qualitatively and quantitatively adaptation capabilities and control accuracy. Results showed improved control accuracy in goaldirected reaching with high noise rejection due to multimodal filtering, and adaptability to dynamical inertial changes, elasticity constraints and human disturbances without the need to relearn the model nor parameter retuning.

show abstract

“…RobCoGen [23] is based on using CppAD [24] along with code generation techniques to get the FO derivatives of Inverse and Forward Dynamics. The open-source symbolic AD toolbox CasADi [19] has been a popular choice in the past for RBD libraries [25], [26] and computing the derivatives of dynamics. CasADi employs forward and reverse mode chain-rule differentiation, and also supports code generation to output compilable C code.…”

Section: Introductionmentioning

confidence: 99%

Analytical Second-Order Partial Derivatives of Rigid-Body Inverse Dynamics

Singh

Russell

Wensing

2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Model-based control for robots has increasingly been dependent on optimization-based methods like Differential Dynamic Programming and iterative LQR (iLQR). These methods can form the basis of Model-Predictive Control (MPC), which is commonly used for controlling legged robots. Computing the partial derivatives of the dynamics is often the most expensive part of these algorithms, regardless of whether analytical methods, Finite Difference, Automatic Differentiation (AD), or Chain-Rule accumulation is used. Since the second-order derivatives of dynamics result in tensor computations, they are often ignored, leading to the use of iLQR, instead of the full second-order DDP method. In this paper, we present analytical methods to compute the second-order derivatives of inverse and forward dynamics for open-chain rigid-body systems with multi-DoF joints and fixed/floating bases. An extensive comparison of accuracy and run-time performance with AD and other methods is provided, including the consideration of code-generation techniques in C/C++ to speed up the computations. For the 36 DoF ATLAS humanoid, the second-order Inverse, and the Forward dynamics derivatives take ≈ 200µs, and ≈ 2.1ms respectively, resulting in a 3× speedup over the AD approach.

show abstract

Robot Dynamics with URDF & CasADi

Cited by 11 publications

References 14 publications

Enhancing fluency and productivity in human-robot collaboration through online scaling of dynamic safety zones

Enhancing fluency and productivity in human-robot collaboration through online scaling of dynamic safety zones

Adaptation through prediction: multisensory active inference torque control

Analytical Second-Order Partial Derivatives of Rigid-Body Inverse Dynamics

Contact Info

Product

Resources

About