A Self-Organizing Neural Model of Motor Equivalent Reaching and Tool Use by a Multijoint Arm

By Daniel Bullock, Stephen Grossberg, Frank H. Guenther. 1993

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Learning to Control in Operational Space (2008, methodology): “Such an approach was first introduced in the context of inverse kinematics learning (Bullock et al. 1993 D’Souza et al. 2001)..”
Toward a new theory of motor synergies. (2007, background): “A less formal, more physiologically based approach has been pursued by Bullock et al. (1993), who proposed a neuronal network that learns coordinate transformations from end-effector to joint coordinates, in effect developing a structure that selects one solution out of a set of " motor-equivalent…”
Self-Organizing Feature Maps for Modeling and Control of Robotic Manipulators (2003, methodology): “This would allow the comparison with other unsupervised models, such as the VITE (Bullock and Grossberg, 1988), the VAM (Gaudiano and Grossberg, 1991) and the DIRECT models (Bullock et al., 1993) in more common grounds.”
Online Goal Babbling for rapid bootstrapping of inverse models in high dimensions (2011, background): “While this redundancy is often considered a problem for sensorimotor learning [7], [8], it also reduces the demand”
Bridging the representational gap in the dynamic systems approach to development (2003, unknown): “The challenge to scale up connectionist models from simple ‘toy’ simulations to simulations that would work with real-world sensors and actuators is probably surmountable, but not trivial (see Bullock & Grossberg, 1988; Bullock et al., 1993 for examples that move in this direction).”
Computational approaches to motor learning by imitation. (2003, unknown): “Task-level imitation requires prior knowledge of how a task-level command can be converted into a command in internal coordinates, a classic problem in control theory treated under the name of inverse kinematics (Baillieul & Martin 1990), but which has found several elegant solutions in neural compu”
Learning inverse kinematics (2001, background): “As noted by Bullock et al. ([ 5 ]), it is possible to transform the non-convex problem of inverse kinematics learning into a convex problem by spatially localizing the learning task: within the vicinity of a particular �, inverse kinematics is actually convex.”
Motor Synergies and the Equilibrium-Point Hypothesis (2010, result): “This model is similar in spirit to an approach advocated by Bullock et al. (1993), who proposed a neuronal network that learns coordinate transformations from end-effector to joint coordinates resulting in a structure that selects one solution out of a set of motor-equivalent solutions.”
Goal Babbling Permits Direct Learning of Inverse Kinematics (2010, methodology): “It has been proposed as an implementation of the “body babbling” introduced by Meltzoff and Moore, but was used also before body babbling was introduced [23], [24].”
Reinforcement learning by reward-weighted regression for operational space control (2007, background): “Nevertheless, the non-convexity issue can be resolved by employing a spatially localized supervised learning system, an approach that was .rst introduced in the context of inverse kinematics learning (Bullock et al., 1993).”