Online motion planning over uneven terrain with walking primitives and regression

Apostolopoulos, Sotiris; Leibold, Marion; Buss, Martin

doi:10.1109/icra.2016.7487568

Cited by 6 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One way of achieving this is to precompute generically useful motion primitives offline, and use them online as a starting point when searching for useful actions (Hauser et al, 2008). Dynamical results have been attained similarly in the lower-dimensional domain of the planar five-link biped (Apostolopoulos et al, 2016).…”

Section: Related Workmentioning

confidence: 99%

Kinodynamic planning for spherical tensegrity locomotion with effective gait primitives

Littlefield

Surovik

Vespignani

et al. 2019

The International Journal of Robotics Research

View full text Add to dashboard Cite

Tensegrity-based robots can achieve locomotion through shape deformation and compliance. They are highly adaptable to their surroundings, and are lightweight, low cost, and physically robust. Their high dimensionality and strongly dynamic nature, however, can complicate motion planning. Efforts to date have primarily considered quasi-static reconfiguration and short-term dynamic motion of tensegrity robots, which do not fully exploit the underlying system dynamics in the long term. Longer-horizon planning has previously required costly search over the full space of valid control inputs. This work synthesizes new and existing approaches to produce dynamic long-term motion while balancing the computational demand. A numerical process based upon quasi-static assumptions is first applied to deform the system into an unstable configuration, causing forward motion. The dynamical characteristics of the result are then altered via a few simple parameters to produce a small but diverse set of useful behaviors. The proposed approach takes advantage of identified symmetries on the prototypical spherical tensegrity robot, which reduce the number of needed gaits but allow motion along different directions. These gaits are first combined with a standard search method to achieve long-term planning in environments where the developed gaits are effective. For more complex environments, the various motion primitives are paired with the fall-back option of random valid actions and are used by an informed sampling-based kinodynamic motion planner with anytime properties. Evaluations using a physics-based model for the prototypical robot demonstrate that modest but efficiently applied search effort can unlock the utility of dynamic tensegrity motion to produce high-quality solutions.

show abstract

Section: Related Workmentioning

confidence: 99%

Kinodynamic planning for spherical tensegrity locomotion with effective gait primitives

Littlefield

Surovik

Vespignani

et al. 2019

The International Journal of Robotics Research

View full text Add to dashboard Cite

show abstract

“…For example, in [21], a finite number of global optimal solutions is computed offline; then, they are generalized using support vector machines or Gaussian process regression and used as training data in the online phase. Similarly, machine learning and motion primitives are used in [22,23] to learn precomputed optimal trajectories. However, in real dynamic environments, an infinite number of cases can occur; thus, it is difficult to cover all possibilities with a finite number of precomputed solutions.…”

Section: Introductionmentioning

confidence: 99%

A Fast and Close-to-Optimal Receding Horizon Control for Trajectory Generation in Dynamic Environments

2022

Self Cite

View full text Add to dashboard Cite

This paper presents a new approach for the optimal trajectory planning of nonlinear systems in a dynamic environment. Given the start and end goals with an objective function, the problem is to find an optimal trajectory from start to end that minimizes the objective while taking into account the changes in the environment. One of the main challenges here is that the optimal control sequence needs to be computed in a limited amount of time and needs to be adapted on-the-fly. The control method presented in this work has two stages: the first-order gradient algorithm is used at the beginning to compute an initial guess of the control sequence that satisfies the constraints but is not yet optimal; then, sequential action control is used to optimize only the portion of the control sequence that will be applied on the system in the next iteration. This helps to reduce the computational effort while still being optimal with regard to the objective; thus, the proposed approach is more applicable for online computation as well as dealing with dynamic environments. We also show that under mild conditions, the proposed controller is asymptotically stable. Different simulated results demonstrate the capability of the controller in terms of solving various tracking problems for different systems under the existence of dynamic obstacles. The proposed method is also compared to the related indirect optimal control approach and sequential action control in terms of cost and computation time to evaluate the improvement of the proposed method.

show abstract

“…21,22 Finally, reference solutions to di®erent task objectives can also be found using Machine Learning techniques following the approach proposed in our previous work. 23 In this previous work however, we focused solely on the feasibility of the task and not in preservation of optimality.…”

Section: Introductionmentioning

confidence: 99%

Energy Efficient and Robust Balancing with Motion Primitive Switching

Apostolopoulos

Leibold

Buss

2017

Int. J. Human. Robot.

Self Cite

View full text Add to dashboard Cite

Balancing motions are usually designed using simpli¯ed models of the Center of Mass (CoM) and feedback control without accounting for energy e±ciency. In order to tackle this shortcoming, we introduce a Motion Primitive switching methodology where samples of optimal motions (Motion Primitives) are chosen online based on a Euclidean distance metric. The chosen sample is used to provide reference trajectories, torques and ground reaction forces to be tracked. In order to satisfy all of the modeling assumptions while tracking the reference values, a Quadratic Program (QP) is solved online where the dynamics of the robot, friction, Center of Pressure and torque bounds are treated as constraints. Convergence to the desired trajectories is dictated by a Control Lyapunov Function constraint which is introduced in the QP. The methodology is evaluated on a four-link simulated robot where we show that switching between Motion Primitives provides energy efcient balancing motions for di®erent disturbance situations. At the same time the methodology provides more e±cient motions for di®erent disturbance forces when compared to a nonswitching approach, where a Motion Primitive is chosen only once at the beginning.

show abstract

Online motion planning over uneven terrain with walking primitives and regression

Cited by 6 publications

References 15 publications

Kinodynamic planning for spherical tensegrity locomotion with effective gait primitives

Kinodynamic planning for spherical tensegrity locomotion with effective gait primitives

A Fast and Close-to-Optimal Receding Horizon Control for Trajectory Generation in Dynamic Environments

Energy Efficient and Robust Balancing with Motion Primitive Switching

Contact Info

Product

Resources

About