An integrated system for real-time model predictive control of humanoid robots

Erez, Tom; Lowrey, Kendall; Tassa, Yuval; Kumar, Vikash; Kolev, Svetoslav; Todorov, Emanuel

doi:10.1109/humanoids.2013.7029990

Cited by 117 publications

(101 citation statements)

References 6 publications

Supporting

Mentioning

101

Contrasting

Order By: Relevance

“…Here we use a first-order version of the Differential Dynamic Programming algorithm [24]. For brevity, here we present only an outline of the algorithm; for more details on our optimization approach, see [7], [8].…”

Section: Online Trajectory Optimizationmentioning

confidence: 99%

Real-time behaviour synthesis for dynamic hand-manipulation

Kumar

Tassa

Erez

et al. 2014

2014 IEEE International Conference on Robotics and Automation (ICRA)

Self Cite

View full text Add to dashboard Cite

Abstract-Dexterous hand manipulation is one of the most complex types of biological movement, and has proven very difficult to replicate in robots. The usual approaches to robotic control -following pre-defined trajectories or planning online with reduced models -are both inapplicable. Dexterous manipulation is so sensitive to small variations in contact force and object location that it seems to require online planning without any simplifications. Here we demonstrate for the first time online planning (or model-predictive control) with a full physics model of a humanoid hand, with 28 degrees of freedom and 48 pneumatic actuators. We augment the actuation space with motor synergies which speed up optimization without removing dexterity. Most of our results are in simulation, showing nonprehensile object manipulation as well as typing. In both cases the input to the system is a high level task description, while all details of the hand movement emerge online from fully automated numerical optimization.

show abstract

Section: Online Trajectory Optimizationmentioning

confidence: 99%

Real-time behaviour synthesis for dynamic hand-manipulation

Kumar

Tassa

Erez

et al. 2014

2014 IEEE International Conference on Robotics and Automation (ICRA)

Self Cite

View full text Add to dashboard Cite

show abstract

“…When using finite difference to calculate Jacobian [1], the compute time increment is negligible. What's more, the optimization could still effectively minimize the total cost (11).…”

Section: A Example: Make a Stridementioning

confidence: 98%

“…For motion planning on a multi-joints humanoid robot with a huge state space, using demonstration movement data becomes important as the solution search space of the motion planning problem could be restricted, making it much easier to solve the problem [1]. Trials are made using whole-body human demonstration to generate motions for a biped robot.…”

Section: Introductionmentioning

confidence: 99%

Follow my step: A framework for biped robots to imitate human walking

Hou

Zhao

2014

2014 IEEE International Conference on Robotics and Biomimetics (ROBIO 2014)

View full text Add to dashboard Cite

This paper presents a motion planning framework to generate continuous walking gait on a biped robot using kinesthetic teaching. While imitating human leg movement, the generated gait could also track a given swing foot trajectory and a given ZMP trajectory simultaneously, so as to form a stable stride with an expected foot displacement. An algorithm based on local linearization and nonlinear optimization is proposed to achieve single foot stance motion planning. An ad-hoc double loop is designed to handle the change of supporting foot so as to form a continuous walking gait with multiple steps. The algorithm is firstly verified via simulation results, then shows its robustness in experiments on the THU-Striker, a 1.34m humanoid robot in robocup2014, Brazil.

show abstract

“…However, in highly cluttered environments, the optimization problem is very highdimensional and enforcing the complementarity constraints arising out of contacts is challenging. Erez et al [8] also use iterative LQR for trajectory optimization for humanoid robots, but this method optimizes over the joint angles and velocities of the robot and does not consider contacts with multiple, dynamic obstacles in the environment.…”

Section: Related Workmentioning

confidence: 98%

Physics-based trajectory optimization for grasping in cluttered environments

Kitaev

Mordatch

Patil

et al. 2015

2015 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Planning motions to grasp an object in cluttered and uncertain environments is a challenging task, particularly when a collision-free trajectory does not exist and objects obstructing the way are required to be carefully grasped and moved out. This paper takes a different approach and proposes to address this problem by using a randomized physics-based motion planner that permits robot-object and object-object interactions. The main idea is to avoid an explicit high-level reasoning of the task by providing the motion planner with a physics engine to evaluate possible complex multi-body dynamical interactions. The approach is able to solve the problem in complex scenarios, also considering uncertainty in the objects' pose and in the contact dynamics. The work enhances the state validity checker, the control sampler and the tree exploration strategy of a kinody-namic motion planner called KPIECE. The enhanced algorithm, called p-KPIECE, has been validated in simulation and with real experiments. The results have been compared with an ontological physics-based motion planner and with task and motion planning approaches, resulting in a significant improvement in terms of planning time, success rate and quality of the solution path.

show abstract

An integrated system for real-time model predictive control of humanoid robots

Cited by 117 publications

References 6 publications

Real-time behaviour synthesis for dynamic hand-manipulation

Real-time behaviour synthesis for dynamic hand-manipulation

Follow my step: A framework for biped robots to imitate human walking

Physics-based trajectory optimization for grasping in cluttered environments

Contact Info

Product

Resources

About