Threading neural feedforward into a mechanical spring: How biology exploits physics in limb control

Kalveram, Karl Theodor; Schinauer, Thomas; Beirle, Steffen; Richter, Stefanie; Jansen-Osmann, Petra

doi:10.1007/s00422-005-0542-6

Cited by 16 publications

(18 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In agreement with predictions from modelling studies, the fixed programme uses the flight time to adapt muscle activity to step height. These adaptations explore system mechanics (Blickhan et al, 2007;Kalveram et al, 2005). In principle, these strategies allow a reduced control effort (Haeufle et al, 2014).…”

Section: Discussionmentioning

confidence: 99%

Preparing the leg for ground contact in running: the contribution of feed-forward and visual feedback

Müller

Häufle

Blickhan

2014

Journal of Experimental Biology

View full text Add to dashboard Cite

While running on uneven ground, humans are able to negotiate visible but also camouflaged changes in ground level. Previous studies have shown that the leg kinematics before touch down change with ground level. The present study experimentally investigated the contributions of visual perception (visual feedback), proprioceptive feedback and feed-forward patterns to the muscle activity responsible for these adaptations. The activity of three bilateral lower limb muscles (m. gastrocnemius medialis, m. tibialis anterior and m. vastus medialis) of nine healthy subjects was recorded during running across visible (drop of 0, −5 and −10 cm) and camouflaged changes in ground level (drop of 0 and −10 cm). The results reveal that at touchdown with longer flight time, m. tibialis anterior activation decreases and m. vastus medialis activation increases purely by feed-forward driven (flight time-dependent) muscle activation patterns, while m. gastrocnemius medialis activation increase is additionally influenced by visual feedback. Thus, feed-forward driven muscle activation patterns are sufficient to explain the experimentally observed adjustments of the leg at touchdown.

show abstract

Section: Discussionmentioning

confidence: 99%

Preparing the leg for ground contact in running: the contribution of feed-forward and visual feedback

Müller

Häufle

Blickhan

2014

Journal of Experimental Biology

View full text Add to dashboard Cite

show abstract

“…It rather considers locomotor movements as emerging from "morphological computation" (Paul 2006), "embodied artificial intelligence" (Pfeifer and Iida 2004), "intelligence by mechanics" (Blickhan et al 2007), or "feedforward preflex control" (Cham et al 2000). We prefer to simply call those processes "exploitive control" (Kalveram et al 2005) or "exploitive actuation" (Kalveram and Seyfarth 2009). This denotation is biologically motivated.…”

Section: Introductionmentioning

confidence: 99%

“…To get knowledge about physical properties of a real system, experimentally administered perturbations are in addition necessary. The experimental method also appears as the tool of choice in making a decision, which of several competing concepts is closer to biological reality (Kalveram et al 2005).…”

Section: Introductionmentioning

confidence: 99%

Energy management that generates terrain following versus apex-preserving hopping in man and machine

et al. 2012

View full text Add to dashboard Cite

While hopping, 12 subjects experienced a sudden step down of 5 or 10 cm. Results revealed that the hopping style was "terrain following". It means that the subjects pursued to keep the distance between maximum hopping height (apex) and ground profile constant. The spring-loaded inverse pendulum (SLIP) model, however, which is currently considered as template for stable legged locomotion would predict apex-preserving hopping, by which the absolute maximal hopping height is kept constant regardless of changes of the ground level. To get more insight into the physics of hopping, we outlined two concepts of energy management: "constant energy supply", by which in each bounce--regardless of perturbations--the same amount of mechanical energy is injected, and "lost energy supply", by which the mechanical energy that is going to be dissipated in the current cycle is assessed and replenished. When tested by simulations and on a robot testbed capable of hopping, constant energy supply generated stable and robust terrain following hopping, whereas lost energy supply led to something like apex-preserving hopping, which, however, lacks stability as well as robustness. Comparing simulated and machine hopping with human hopping suggests that constant energy supply has a good chance to be used by humans to generate hopping.

show abstract

“…This also means that, before the learning process is completed, the motor system works in an open-loop and any perturbation could easily destabilize the body-plant control scheme. This undesirable situation has been traditionally circumvented by adding a sensory feedback control loop accompanied on most occasions with some kind of feedback controller (Kawato and Gomi, 1992; Stroeve, 1997; Desmurget and Grafton, 2000; Kalveram et al, 2005). This latter ranges from a simple proportional—derivative control (Van Der Smagt, 2000), which ensures stability (Arimoto, 1984), to other more sophisticated controllers following the general structure shown in Figure 1A.…”

Section: Introductionmentioning

confidence: 99%

Fast convergence of learning requires plasticity between inferior olive and deep cerebellar nuclei in a manipulation task: a closed-loop robotic simulation

Luque

Garrido

Carrillo

et al. 2014

Front. Comput. Neurosci.

View full text Add to dashboard Cite

The cerebellum is known to play a critical role in learning relevant patterns of activity for adaptive motor control, but the underlying network mechanisms are only partly understood. The classical long-term synaptic plasticity between parallel fibers (PFs) and Purkinje cells (PCs), which is driven by the inferior olive (IO), can only account for limited aspects of learning. Recently, the role of additional forms of plasticity in the granular layer, molecular layer and deep cerebellar nuclei (DCN) has been considered. In particular, learning at DCN synapses allows for generalization, but convergence to a stable state requires hundreds of repetitions. In this paper we have explored the putative role of the IO-DCN connection by endowing it with adaptable weights and exploring its implications in a closed-loop robotic manipulation task. Our results show that IO-DCN plasticity accelerates convergence of learning by up to two orders of magnitude without conflicting with the generalization properties conferred by DCN plasticity. Thus, this model suggests that multiple distributed learning mechanisms provide a key for explaining the complex properties of procedural learning and open up new experimental questions for synaptic plasticity in the cerebellar network.

show abstract

Threading neural feedforward into a mechanical spring: How biology exploits physics in limb control

Cited by 16 publications

References 40 publications

Preparing the leg for ground contact in running: the contribution of feed-forward and visual feedback

Preparing the leg for ground contact in running: the contribution of feed-forward and visual feedback

Energy management that generates terrain following versus apex-preserving hopping in man and machine

Fast convergence of learning requires plasticity between inferior olive and deep cerebellar nuclei in a manipulation task: a closed-loop robotic simulation

Contact Info

Product

Resources

About