Limits...
Distributed recurrent neural forward models with synaptic adaptation and CPG-based control for complex behaviors of walking robots.

Dasgupta S, Goldschmidt D, Wörgötter F, Manoonpong P - Front Neurorobot (2015)

Bottom Line: Biological study has revealed that such complex behaviors are a result of a combination of biomechanics and neural mechanism thus representing the true nature of embodied interactions.Inspired by these findings, we present here, an artificial bio-inspired walking system which effectively combines biomechanics (in terms of the body and leg structures) with the underlying neural mechanisms.Furthermore, we demonstrate that the newly developed recurrent network based approach to online forward models outperforms the adaptive neuron forward models, which have hitherto been the state of the art, to model a subset of similar walking behaviors in walking robots.

View Article: PubMed Central - PubMed

Affiliation: Institute for Physics - Biophysics, George-August-University Göttingen, Germany ; Bernstein Center for Computational Neuroscience, George-August-University Göttingen, Germany ; Laboratory for Neural Computation and Adaptation, Riken Brain Science Institute Saitama, Japan.

ABSTRACT
Walking animals, like stick insects, cockroaches or ants, demonstrate a fascinating range of locomotive abilities and complex behaviors. The locomotive behaviors can consist of a variety of walking patterns along with adaptation that allow the animals to deal with changes in environmental conditions, like uneven terrains, gaps, obstacles etc. Biological study has revealed that such complex behaviors are a result of a combination of biomechanics and neural mechanism thus representing the true nature of embodied interactions. While the biomechanics helps maintain flexibility and sustain a variety of movements, the neural mechanisms generate movements while making appropriate predictions crucial for achieving adaptation. Such predictions or planning ahead can be achieved by way of internal models that are grounded in the overall behavior of the animal. Inspired by these findings, we present here, an artificial bio-inspired walking system which effectively combines biomechanics (in terms of the body and leg structures) with the underlying neural mechanisms. The neural mechanisms consist of (1) central pattern generator based control for generating basic rhythmic patterns and coordinated movements, (2) distributed (at each leg) recurrent neural network based adaptive forward models with efference copies as internal models for sensory predictions and instantaneous state estimations, and (3) searching and elevation control for adapting the movement of an individual leg to deal with different environmental conditions. Using simulations we show that this bio-inspired approach with adaptive internal models allows the walking robot to perform complex locomotive behaviors as observed in insects, including walking on undulated terrains, crossing large gaps, leg damage adaptations, as well as climbing over high obstacles. Furthermore, we demonstrate that the newly developed recurrent network based approach to online forward models outperforms the adaptive neuron forward models, which have hitherto been the state of the art, to model a subset of similar walking behaviors in walking robots.

No MeSH data available.


Related in: MedlinePlus

(A–C) The CTr-joint motor signal of the right front leg (R1) for wave, tetrapod, and caterpillar gaits, respectively. This motor signal provides the efference copy or the input to the reservoir forward models. (D–F) The actual foot contact signal (force sensor signal under normal walking conditions) used as the target signal of the reservoir models. (G–I) The predicted foot contact signal or the final learned output of the forward model for each walking gait (RF output signal). The green shaded region indicates the time interval between swing and stance phase for the CTr motor signal at the three walking gaits. As observed the actual foot contact signal is considerably lagged in time compared to the motor signal. Effectively, this lag decreases with an increase in the gait frequency. The single RF adaptively accounts for these different delay times in order to accurately predict the expected foot contact signal. (J) above—CTr-joint motor signal demonstrated for a single leg, with 2% Gaussian noise injected between 300 and 350 time steps (yellow shaded region), below—Despite the noise corruption of the motor signal, the reservoir forward model is able to generate the correct predicted FC signal (blue dotted—target FC signal, red solid—predicted signal). (K) above—The CTr-joint motor signal corrupted with missing information between 280 and 320 time steps. As a result, the motor signal shows a narrow spike between 310 and 330 time steps (yellow shaded region), below—Reservoir forward model predicted signal (red) as compared to the desired FC signal (dotted blue). Although the CTr motor signal was transiently missing, the reservoir is able to generate the desired FC signal considerably well, while at the same time maintaining the correct temporal sequence of the signals.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4585172&req=5

Figure 5: (A–C) The CTr-joint motor signal of the right front leg (R1) for wave, tetrapod, and caterpillar gaits, respectively. This motor signal provides the efference copy or the input to the reservoir forward models. (D–F) The actual foot contact signal (force sensor signal under normal walking conditions) used as the target signal of the reservoir models. (G–I) The predicted foot contact signal or the final learned output of the forward model for each walking gait (RF output signal). The green shaded region indicates the time interval between swing and stance phase for the CTr motor signal at the three walking gaits. As observed the actual foot contact signal is considerably lagged in time compared to the motor signal. Effectively, this lag decreases with an increase in the gait frequency. The single RF adaptively accounts for these different delay times in order to accurately predict the expected foot contact signal. (J) above—CTr-joint motor signal demonstrated for a single leg, with 2% Gaussian noise injected between 300 and 350 time steps (yellow shaded region), below—Despite the noise corruption of the motor signal, the reservoir forward model is able to generate the correct predicted FC signal (blue dotted—target FC signal, red solid—predicted signal). (K) above—The CTr-joint motor signal corrupted with missing information between 280 and 320 time steps. As a result, the motor signal shows a narrow spike between 310 and 330 time steps (yellow shaded region), below—Reservoir forward model predicted signal (red) as compared to the desired FC signal (dotted blue). Although the CTr motor signal was transiently missing, the reservoir is able to generate the desired FC signal considerably well, while at the same time maintaining the correct temporal sequence of the signals.

Mentions: Reservoir-to-readout weight adaptation during online learning. (A) Changes of 30 weights projecting to the first readout neuron (z1) of the forward model of the right front leg (R1) while walking with a wave gait. During this period, weights projecting to the second (z2) and third (z3) output neurons remain unchanged (i.e., they are zero). (B) Changes of the weights to z2 while walking with a tetrapod gait. During this period, the weights to z3 still remain unchanged and the weights to z1 converge to around zero. (C) Changes of the weights to z3 while walking with a caterpillar gait. During this period, the weights to z1 and z2 converge to around zero. At the end of each gait, all weights are stored such that they will be used for locomotion in different environments. The gray areas represent transition phases from one gait to another gait and the yellow areas represent convergence. The gait diagrams are shown on the right. They are observed from the motor signals of the CTr-joints (Figure 5). White areas indicate ground contact or stance phase and blue areas refer to no ground contact during swing phase. As frequency increases, some legs step in pairs (dashed enclosures). Here convergence implies no significant change in the vector norm of the readout weights.


Distributed recurrent neural forward models with synaptic adaptation and CPG-based control for complex behaviors of walking robots.

Dasgupta S, Goldschmidt D, Wörgötter F, Manoonpong P - Front Neurorobot (2015)

(A–C) The CTr-joint motor signal of the right front leg (R1) for wave, tetrapod, and caterpillar gaits, respectively. This motor signal provides the efference copy or the input to the reservoir forward models. (D–F) The actual foot contact signal (force sensor signal under normal walking conditions) used as the target signal of the reservoir models. (G–I) The predicted foot contact signal or the final learned output of the forward model for each walking gait (RF output signal). The green shaded region indicates the time interval between swing and stance phase for the CTr motor signal at the three walking gaits. As observed the actual foot contact signal is considerably lagged in time compared to the motor signal. Effectively, this lag decreases with an increase in the gait frequency. The single RF adaptively accounts for these different delay times in order to accurately predict the expected foot contact signal. (J) above—CTr-joint motor signal demonstrated for a single leg, with 2% Gaussian noise injected between 300 and 350 time steps (yellow shaded region), below—Despite the noise corruption of the motor signal, the reservoir forward model is able to generate the correct predicted FC signal (blue dotted—target FC signal, red solid—predicted signal). (K) above—The CTr-joint motor signal corrupted with missing information between 280 and 320 time steps. As a result, the motor signal shows a narrow spike between 310 and 330 time steps (yellow shaded region), below—Reservoir forward model predicted signal (red) as compared to the desired FC signal (dotted blue). Although the CTr motor signal was transiently missing, the reservoir is able to generate the desired FC signal considerably well, while at the same time maintaining the correct temporal sequence of the signals.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4585172&req=5

Figure 5: (A–C) The CTr-joint motor signal of the right front leg (R1) for wave, tetrapod, and caterpillar gaits, respectively. This motor signal provides the efference copy or the input to the reservoir forward models. (D–F) The actual foot contact signal (force sensor signal under normal walking conditions) used as the target signal of the reservoir models. (G–I) The predicted foot contact signal or the final learned output of the forward model for each walking gait (RF output signal). The green shaded region indicates the time interval between swing and stance phase for the CTr motor signal at the three walking gaits. As observed the actual foot contact signal is considerably lagged in time compared to the motor signal. Effectively, this lag decreases with an increase in the gait frequency. The single RF adaptively accounts for these different delay times in order to accurately predict the expected foot contact signal. (J) above—CTr-joint motor signal demonstrated for a single leg, with 2% Gaussian noise injected between 300 and 350 time steps (yellow shaded region), below—Despite the noise corruption of the motor signal, the reservoir forward model is able to generate the correct predicted FC signal (blue dotted—target FC signal, red solid—predicted signal). (K) above—The CTr-joint motor signal corrupted with missing information between 280 and 320 time steps. As a result, the motor signal shows a narrow spike between 310 and 330 time steps (yellow shaded region), below—Reservoir forward model predicted signal (red) as compared to the desired FC signal (dotted blue). Although the CTr motor signal was transiently missing, the reservoir is able to generate the desired FC signal considerably well, while at the same time maintaining the correct temporal sequence of the signals.
Mentions: Reservoir-to-readout weight adaptation during online learning. (A) Changes of 30 weights projecting to the first readout neuron (z1) of the forward model of the right front leg (R1) while walking with a wave gait. During this period, weights projecting to the second (z2) and third (z3) output neurons remain unchanged (i.e., they are zero). (B) Changes of the weights to z2 while walking with a tetrapod gait. During this period, the weights to z3 still remain unchanged and the weights to z1 converge to around zero. (C) Changes of the weights to z3 while walking with a caterpillar gait. During this period, the weights to z1 and z2 converge to around zero. At the end of each gait, all weights are stored such that they will be used for locomotion in different environments. The gray areas represent transition phases from one gait to another gait and the yellow areas represent convergence. The gait diagrams are shown on the right. They are observed from the motor signals of the CTr-joints (Figure 5). White areas indicate ground contact or stance phase and blue areas refer to no ground contact during swing phase. As frequency increases, some legs step in pairs (dashed enclosures). Here convergence implies no significant change in the vector norm of the readout weights.

Bottom Line: Biological study has revealed that such complex behaviors are a result of a combination of biomechanics and neural mechanism thus representing the true nature of embodied interactions.Inspired by these findings, we present here, an artificial bio-inspired walking system which effectively combines biomechanics (in terms of the body and leg structures) with the underlying neural mechanisms.Furthermore, we demonstrate that the newly developed recurrent network based approach to online forward models outperforms the adaptive neuron forward models, which have hitherto been the state of the art, to model a subset of similar walking behaviors in walking robots.

View Article: PubMed Central - PubMed

Affiliation: Institute for Physics - Biophysics, George-August-University Göttingen, Germany ; Bernstein Center for Computational Neuroscience, George-August-University Göttingen, Germany ; Laboratory for Neural Computation and Adaptation, Riken Brain Science Institute Saitama, Japan.

ABSTRACT
Walking animals, like stick insects, cockroaches or ants, demonstrate a fascinating range of locomotive abilities and complex behaviors. The locomotive behaviors can consist of a variety of walking patterns along with adaptation that allow the animals to deal with changes in environmental conditions, like uneven terrains, gaps, obstacles etc. Biological study has revealed that such complex behaviors are a result of a combination of biomechanics and neural mechanism thus representing the true nature of embodied interactions. While the biomechanics helps maintain flexibility and sustain a variety of movements, the neural mechanisms generate movements while making appropriate predictions crucial for achieving adaptation. Such predictions or planning ahead can be achieved by way of internal models that are grounded in the overall behavior of the animal. Inspired by these findings, we present here, an artificial bio-inspired walking system which effectively combines biomechanics (in terms of the body and leg structures) with the underlying neural mechanisms. The neural mechanisms consist of (1) central pattern generator based control for generating basic rhythmic patterns and coordinated movements, (2) distributed (at each leg) recurrent neural network based adaptive forward models with efference copies as internal models for sensory predictions and instantaneous state estimations, and (3) searching and elevation control for adapting the movement of an individual leg to deal with different environmental conditions. Using simulations we show that this bio-inspired approach with adaptive internal models allows the walking robot to perform complex locomotive behaviors as observed in insects, including walking on undulated terrains, crossing large gaps, leg damage adaptations, as well as climbing over high obstacles. Furthermore, we demonstrate that the newly developed recurrent network based approach to online forward models outperforms the adaptive neuron forward models, which have hitherto been the state of the art, to model a subset of similar walking behaviors in walking robots.

No MeSH data available.


Related in: MedlinePlus