Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userInverse Dynamics vs. Forward Dynamics in Direct Transcription Formulations for Trajectory Optimization
206
0
0.0
(
0
)
Added by
Henrique Ferrolho
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Benchmarks of state-of-the-art rigid-body dynamics libraries report better performance solving the inverse dynamics problem than the forward alternative. Those benchmarks encouraged us to question whether that computational advantage would translate to direct transcription, where calculating rigid-body dynamics and their derivatives accounts for a significant share of computation time. In this work, we implement an optimization framework where both approaches for enforcing the system dynamics are available. We evaluate the performance of each approach for systems of varying complexity, for domains with rigid contacts. Our tests reveal that formulations using inverse dynamics converge faster, require less iterations, and are more robust to coarse problem discretization. These results indicate that inverse dynamics should be preferred to enforce the nonlinear system dynamics in simultaneous methods, such as direct transcription.
rate research
Read More
We present a framework for bi-level trajectory optimization in which a systems dynamics are encoded as the solution to a constrained optimization problem and smooth gradients of this lower-level problem are passed to an upper-level trajectory optimizer. This optimization-based dynamics representation enables constraint handling, additional variables, and non-smooth forces to be abstracted away from the upper-level optimizer, and allows classical unconstrained optimizers to synthesize trajectories for more complex systems. We provide a path-following method for efficient evaluation of constrained dynamics and utilize the implicit-function theorem to compute smooth gradients of this representation. We demonstrate the framework by modeling systems from locomotion, aerospace, and manipulation domains including: acrobot with joint limits, cart-pole subject to Coulomb friction, Raibert hopper, rocket landing with thrust limits, and planar-push task with optimization-based dynamics and then optimize trajectories using iterative LQR.
Estimating accurate forward and inverse dynamics models is a crucial component of model-based control for sophisticated robots such as robots driven by hydraulics, artificial muscles, or robots dealing with different contact situations. Analytic models to such processes are often unavailable or inaccurate due to complex hysteresis effects, unmodelled friction and stiction phenomena,and unknown effects during contact situations. A promising approach is to obtain spatio-temporal models in a data-driven way using recurrent neural networks, as they can overcome those issues. However, such models often do not meet accuracy demands sufficiently, degenerate in performance for the required high sampling frequencies and cannot provide uncertainty estimates. We adopt a recent probabilistic recurrent neural network architecture, called Re-current Kalman Networks (RKNs), to model learning by conditioning its transition dynamics on the control actions. RKNs outperform standard recurrent networks such as LSTMs on many state estimation tasks. Inspired by Kalman filters, the RKN provides an elegant way to achieve action conditioning within its recurrent cell by leveraging additive interactions between the current latent state and the action variables. We present two architectures, one for forward model learning and one for inverse model learning. Both architectures significantly outperform exist-ing model learning frameworks as well as analytical models in terms of prediction performance on a variety of real robot dynamics models.
Simplified models are useful to increase the computational efficiency of a motion planning algorithm, but their lack of accuracy have to be managed. We propose two feasibility constraints to be included in a Single Rigid Body Dynamicsbased trajectory optimizer in order to obtain robust motions in challenging terrain. The first one finds an approximate relationship between joint-torque limits and admissible contact forces, without requiring the joint positions. The second one proposes a leg model to prevent leg collision with the environment. Such constraints have been included in a simplified nonlinear nonconvex trajectory optimization problem. We demonstrate the feasibility of the resulting motion plans both in simulation and on the Hydraulically actuated Quadruped (HyQ) robot, considering experiments on an irregular terrain.
Being able to quickly adapt to changes in dynamics is paramount in model-based control for object manipulation tasks. In order to influence fast adaptation of the inverse dynamics models parameters, data efficiency is crucial. Given observed data, a key element to how an optimizer updates model parameters is the loss function. In this work, we propose to apply meta-learning to learn structured, state-dependent loss functions during a meta-training phase. We then replace standard losses with our learned losses during online adaptation tasks. We evaluate our proposed approach on inverse dynamics learning tasks, both in simulation and on real hardware data. In both settings, the structured and state-dependent learned losses improve online adaptation speed, when compared to standard, state-independent loss functions.
We demonstrate an efficient algorithm for inverse problems in time-dependent quantum dynamics based on feedback loops between Hamiltonian parameters and the solutions of the Schr{o}dinger equation. Our approach formulates the inverse problem as a target vector estimation problem and uses Bayesian surrogate models of the Schr{o}dinger equation solutions to direct the optimization of feedback loops. For the surrogate models, we use Gaussian processes with vector outputs and composite kernels built by an iterative algorithm with Bayesian information criterion (BIC) as a kernel selection metric. The outputs of the Gaussian processes are designed to model an observable simultaneously at different time instances. We show that the use of Gaussian processes with vector outputs and the BIC-directed kernel construction reduce the number of iterations in the feedback loops by, at least, a factor of 3. We also demonstrate an application of Bayesian optimization for inverse problems with noisy data. To demonstrate the algorithm, we consider the orientation and alignment of polyatomic molecules SO$_2$ and chiral propylene oxide (PPO) induced by strong laser pulses. We use simulated time evolutions of the orientation or alignment signals to determine the relevant components of the molecular polarizability tensors to within 1% accuracy. We show that, for the five independent components of the polarizability tensor of PPO, this can be achieved with as few as 30 quantum dynamics calculations.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
