Studying the Effect of Different Optimization Criteria on Humanoid Walking Motions

Koch, Kai Henning; Mombaur, Katja; Souères, Philippe

doi:10.1007/978-3-642-34327-8_22

Cited by 13 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The criteria used in motion imitation other than motion similarity in terms of joint angles [16] include minimum consumed energy (MCE) [41,42,43], minimum torque change (MTC) [44], maximum motion velocity (MMV) [45] and minimum joint velocities (MJV) [46]. The advantages of MCE criterion are that the generated motion was relatively similar to that of human; and battery life or operation time is long.…”

Section: ) Similarity Between Human Motion and Robot Motionmentioning

confidence: 99%

“…The advantage of MTC criterion is that smooth variation in link acceleration and torque results in more stable generated motion [44]. MMV criterion was reported to produce gaits with increased speed but also required an additional constraint on foot impacts; and MJV criterion was reported to produce unnatural and non-smooth gaits with backward tilting of pelvis and fluctuation in pelvis height [46]. In the proposed method, motion similarity (MS) in terms of joint angles is used as the only criterion in the cost function because the implementation of the MS criterion is straightforward due to availability of joint angle trajectories; and it does not require derivation of forces acted on joints and joint torques as in MCE criterion.…”

Section: ) Similarity Between Human Motion and Robot Motionmentioning

confidence: 99%

See 1 more Smart Citation

Imitation of Dynamic Walking with BSN for a Humanoid Robot

Teachasrisaksakul

Zhang

Yang

et al. 2015

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Section: ) Similarity Between Human Motion and Robot Motionmentioning

confidence: 99%

Section: ) Similarity Between Human Motion and Robot Motionmentioning

confidence: 99%

Imitation of Dynamic Walking with BSN for a Humanoid Robot

Teachasrisaksakul

Zhang

Yang

et al. 2015

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

“…Optimal control has previously been used to generate challenging motions using whole-body models for robots such as HRP-2 [12][13][14][15][16][17][18]. In an approach that is situated between whole-body optimization and motion generation by reduced models, optimization of template models has been used for generating walking motion of humanoid robots in challenging scenarios [19].…”

Section: Introductionmentioning

confidence: 99%

Bio-Inspired Optimal Control Framework to Generate Walking Motions for the Humanoid Robot iCub Using Whole Body Models

Mombaur

2018

Applied Sciences

View full text Add to dashboard Cite

Bipedal locomotion remains one of the major open challenges of humanoid robotics. The common approaches are based on simple reduced model dynamics to generate walking trajectories, often neglecting the whole-body dynamics of the robots. As motions in nature are often considered as optimal with respect to certain criteria, in this work, we present an optimal control-based approach that allows us to generate optimized walking motions using a precise whole-body dynamic model of the robot, in contrast with the common approaches. The optimal control problem is formulated to minimize a set of desired objective functions with respect to physical constraints of the robot and contact constraints of the walking phases; the problem is then solved with a direct multiple shooting method. We apply the formulation with combinations of different objective criteria to the model of a reduced version of the iCub humanoid robot of 15 internal DOF. The obtained trajectories are executed on the real robot, and we carry out a discussion on the differences between the outcomes of this approach with the classic approaches.

show abstract

“…Optimal control has previously been used to generate challenging motions using whole-body models for robots such as HRP-2 [9][10][11][12][13][14][15]. In an approach which is situated between whole-body optimization and motion generation by reduced models, optimization of template models has been used for generating walking motion of humanoid robots in challenging scenarios [16].…”

Section: Introductionmentioning

confidence: 99%

Bio-Inspired Optimal Control Framework to Generate Walking Motions for the Humanoid Robot iCub Using Whole Body Models

Hu¹,

Mombaur²

2017

Preprint

View full text Add to dashboard Cite

as optimal with respect to certain criteria, in this work we present an optimal control based approach 4 that allows us to generate optimized walking motions using a precise whole-body dynamic model 5 of the robot, in contrast with the common approaches. The optimal control problem is formulated 6 to minimize a set of desired objective functions with respect to physical constraints of the robot 7 and contact constraints of the walking phases. We apply the method with combinations of different 8 objective criteria to the model of a reduced version of the iCub humanoid robot of 15 internal DOF. The 9 obtained trajectories are executed on the real robot and we carry out a discussion on the differences 10 between the outcomes of this approach with the classic approaches.

show abstract

Studying the Effect of Different Optimization Criteria on Humanoid Walking Motions

Cited by 13 publications

References 14 publications

Imitation of Dynamic Walking with BSN for a Humanoid Robot

Imitation of Dynamic Walking with BSN for a Humanoid Robot

Bio-Inspired Optimal Control Framework to Generate Walking Motions for the Humanoid Robot iCub Using Whole Body Models

Bio-Inspired Optimal Control Framework to Generate Walking Motions for the Humanoid Robot iCub Using Whole Body Models

Contact Info

Product

Resources

About