Billard, Learning motor skills by imitation: a biologically inspired robotic model, Cybernetics & Systems, 32(1-2), 2001, 155-193.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... critic as follows: a) At the onset of the stimulus: x l1 = 1; xl;m6=1 = 0, b) Otherwise, k steps after the onset of the stimulus: xl;mk = 0; xl;m k ae m Gamma1 , where l is the stimulus, m is the component of its corresponding temporal representation, and ae is a constant value in the range [0 1]. 3. Select an action an(t) based on the weighted sum of the stimuli traces and a uniformly distributed source of noise oe n (t) 2 [0; oe] an (t) P l v nl e l (t) Gamma oe n(t) A winner take all rule prevents the system from performing two actions at the same time. 4. The critic ....

....nl (t Gamma 1) jar(t)an(t)e l (t) where, ja is the learning rate of the actor. 7. The action and stimuli traces are updated as follows: an(t) min(1; an(t) ffi an(t Gamma 1) and e l (t) min(1; e l (t Gamma 2) ffi e l (t Gamma 1) where ffi is a constant value in the range [0 1]. 8. Goto step 1. Fig. 2. The inner loop of the TD learning algorithm associated to the time delay actorcritic neural architecture with dopamine like reinforcement signal (adapted from [10] able to associate a reward delivered up to 3 seconds after the selection of an action. 4 Spatial choice ....

[Article contains additional citation context not shown here]

A. Billard. Learning motor skills by imitation: a biologically inspired robotic model. Cybernetics and Systems, 32(1-2), 2001 (in press).

....profile in macaques also showed significantly more variation than that in humans (Fig. 6) Faster speed of movement in macaques is (partly at least) due to their having lighter limbs and greater muscular strength (relative to their body mass) than humans [8] as shown by simulation studies [4,5]. As mentioned earlier in the text, macaques shoulderjoint complex appears to allow less freedom of movements than does the human s one. Confirming this assumption, the natural prehension data show that, in contrast to humans, macaques make smaller shoulder abduction and wider elbow excursion ....

....in Fig. 8) Moreover, the two experimental set ups put similar constraints for the hand motion, as the distance object mouth (when scaled to arm length) was the same for both monkeys and humans and was contained within the same overall space along the vertical and horizontal axes (Fig. 1) In Ref. [4], we carried out computer simulations to reproduce the present experiments. Simulation meant to tell apart the part played by the physics in the differences observed in the two species behaviors. The simulations showed that it was physically possible for macaques to reach all objects by performing ....

[Article contains additional citation context not shown here]

Billard A. Learning motor skills by imitation: a biologically inspired robotic model. Cybernet Syst 2001;32:1 -- 2.

....body of 3 research in robotics, and contribute to data segmentation and understanding. However, they provide highly task specific solutions, with little flexibility for applying the same algorithm to imitation after different types of movements and tasks. More recent efforts, including our own [3, 8, 10, 37], have been oriented toward analyzing the underlying mechanisms of imitation in natural systems and modeling those on artificial ones. Atkeson and Schaal [5, 49] developed a control strategy in which the robot learns a reward function from the demonstration and a task model from repeated attempts ....

....of the model and the dynamic simulation at modeling human imitation. In the experiments presented here, only 11 DOFs are actively commanded to match the observed performance (4 DOFs per arm and 3 for the torso) while the rest of the joints are kept immobile. In the experiments reported in [8, 10], we demonstrated the validity of the architecture for controlling the 65 DOFs of our avatar for imitating complex movements requiring all limbs. There, data for the imitation were simulated, produced by a demonstrator avatar, and we could generate data for the whole body. In this paper, we use ....

A. Billard and M. Mataric. Learning motor skills by imitation: a biologically inspired robotic model. In Fourth International Conference on Autonomous Agents (Agents 2000), Barcelona, Catalonia, Spain. June 3 - June 7, 2000.

....body of 3 research in robotics, and contribute to data segmentation and understanding. However, they provide highly task specific solutions, with little flexibility for applying the same algorithm to imitation after different types of movements and tasks. More recent efforts, including our own [3, 8, 10, 37], have been oriented toward analyzing the underlying mechanisms of imitation in natural systems and modeling those on artificial ones. Atkeson and Schaal [5, 49] developed a control strategy in which the robot learns a reward function from the demonstration and a task model from repeated attempts ....

....structure underlying primates 4 brain s visuo motor pathways. These are the spinal cord, the primary and pre motor cortices (M1 PM) the cerebellum and the temporal cortex. The model has first been evaluated in a pair of demonstrator imitator humanoid avatars with 65 degrees of freedom [8] for learning by imitation gestures and complex movements involving all the avatar s limbs. In this paper, we evaluate the model s performance at reproducing human arm movements. A biomechanical simulation is developed which models the muscles and the complete dynamics of a 37 degree of freedom ....

[Article contains additional citation context not shown here]

A. Billard. Learning motor skills by imitation: a biologically inspired robotic model. Cybernetics & Systems Journal, special issue on Imitation in animals and artifacts, 32:1-2:155-- 193, 2001.

....learning or adaptive capabilities, which can be used for on and or o line optimization of prede ned motor control parameters. Recent e orts, including our own, have been oriented toward analyzing the underlying mechanisms of imitation in natural systems and modeling those on arti cial ones [3, 5, 4, 7, 17, 23]. The endeavor, there, is, on the one hand, to build biologically plausible models of animal s imitative abilities, and, on the other hand, to develop architecture for visuo motor control and learning in robots which would show some of the exibility of natural systems. Our work wishes to ....

....movements and oscillatory movements of the two arms. We compare the model s performance to that of humans in the same imitation task. Section 4 concludes this paper with a short summary of the presented work. 2 The model We have developed a highly simpli ed model of primate imitative ability [5] (see Figure 1) This model 1 Attentional module Inhibit learning until observe a change in one limb position Visual System Learning System Cerebellum module Decision module Learning of coordinated and of spinal networks Predined movements for reaching and grasping activation of nodes ....

[Article contains additional citation context not shown here]

A. Billard. Learning motor skills by imitation: a biologically inspired robotic model. Cybernetics & Systems Journal, special issue on Imitation in animals and artifacts, 2000. To appear.

....solutions, with little flexibility for applying the same algorithm to imitation after types of movements and tasks. More recent efforts, including our own, have been oriented toward analyzing the underlying mechanisms of imitation in natural systems and modeling those on artificial ones [4, 6, 5, 9, 10, 31, 42, 41]. The endeavor, there, is, on the one hand, to build biologically plausible models of animal s imitative abilities, and, on the other hand, to develop architecture for visuo motor control and learning in robots which would show some of the flexibility of natural systems. 2 Our work wishes to ....

....neurological structure underlying primates brain s visuo motor pathways. These are the spinal cord, the primary and pre motor cortexes (M1 PM) the cerebellum and the temporal cortex. The model has first been evaluated in a pair of demonstrator imitator humanoid avatars with 65 degrees of freedom[6] for learning by imitation gestures and complex movements involving all the avatar s limbs. In this paper, we evaluate the model s performance at reproducing human arm movements. A biomechanical simulation is developed which models the muscles and the complete dynamics of a 37 degree of freedom ....

[Article contains additional citation context not shown here]

A. Billard. Learning motor skills by imitation: a biologically inspired robotic model. Cybernetics & Systems Journal, special issue on Imitation in animals and artifacts, 2000. To appear.

No context found.

Billard, Learning motor skills by imitation: a biologically inspired robotic model, Cybernetics & Systems, 32(1-2), 2001, 155-193.

No context found.

Aude Billard. Learning motor skills by imitation: A biologically inspired robotic model. Cybernetics and Systems, 32:155--193, 2001.

No context found.

A. Billard. Learning motor skills by imitation: a biologically inspired robotic model. In Cybernetics and Systems, volume 32:1-2, pages 155--193, 2000.

No context found.

Billard, A. (2000). Learning motor skills by imitation: a biologically inspired robotic model. Cybernetics & Systems.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC