Abstract:Neuromorphic implementation of robotic control has been shown to outperform conventional control paradigms in terms of robustness to perturbations and adaptation to varying conditions. Two main ingredients of robotics are inverse kinematic and Proportional–Integral–Derivative (PID) control. Inverse kinematics is used to compute an appropriate state in a robot's configuration space, given a target position in task space. PID control applies responsive correction signals to a robot's actuators, allowing it to re… Show more
“…The network architecture is discussed in detail in Zaidel et al. 11 Error convergence and reaching point [0.17, 0.17, 0.35] are demonstrated in Figures 4 E and 4F. The comparison between learning and SGD-recurrent implementation is shown in Figure 4 G. Figure 4 Learning and recurrent SNNs for inverse kinematics (A) Simplified recurrent SNN for IK.…”
Section: Resultsmentioning
confidence: 99%
“…Another approach would be using a learning-based derivation of IK via the PES learning rule (Figure 4D). The network architecture is discussed in detail in Zaidel et al 11 Error convergence and reaching point [0.17, 0.17, 0.35] are demonstrated in Figures 4E and 4F. The comparison between learning and SGDrecurrent implementation is shown in Figure 4G.…”
Section: Recurrent and Learning-based Snnmentioning
confidence: 96%
“…The simulated robotic arm model has six DoF, out of which, here, we controlled five (the sixth DoF accounts for gripping). The model follows the physical arm design described in Zaidel et al 11 and is visualized in Figure 1. The arm features six links l 1À6 (12-, 30-, 6-, 20-, 10-, and 20-cm long, respectively) and five joints q 1À5 (rotated around the y, x, x, z, x axes, respectively).…”
Section: Robotic Modelmentioning
confidence: 99%
“…NEF is extensively used to design neuromorphic systems capable of visual perception 10 and motor control. 11 It serves as the foundation for Nengo, a Python-based "neural compiler," which translates high-level descriptions to low-level neural models. 12 NEF-inspired neuromorphic hardware designs 13 have been implemented in both analog and digital circuitry.…”
Summary
Inverse kinematics is fundamental for computational motion planning. It is used to derive an appropriate state in a robot's configuration space, given a target position in task space. In this work, we investigate the performance of fully connected and residual artificial neural networks as well as recurrent, learning-based, and deep spiking neural networks for conventional and geometrically constrained inverse kinematics. We show that while highly parameterized data-driven neural networks with tens to hundreds of thousands of parameters exhibit sub-ms inference time and sub-mm accuracy, learning-based spiking architectures can provide reasonably good results with merely a few thousand neurons. Moreover, we show that spiking neural networks can perform well in geometrically constrained task space, even when configured to an energy-conserved spiking rate, demonstrating their robustness. Neural networks were evaluated on NVIDIA's Xavier and Intel's neuromorphic Loihi chip.
“…The network architecture is discussed in detail in Zaidel et al. 11 Error convergence and reaching point [0.17, 0.17, 0.35] are demonstrated in Figures 4 E and 4F. The comparison between learning and SGD-recurrent implementation is shown in Figure 4 G. Figure 4 Learning and recurrent SNNs for inverse kinematics (A) Simplified recurrent SNN for IK.…”
Section: Resultsmentioning
confidence: 99%
“…Another approach would be using a learning-based derivation of IK via the PES learning rule (Figure 4D). The network architecture is discussed in detail in Zaidel et al 11 Error convergence and reaching point [0.17, 0.17, 0.35] are demonstrated in Figures 4E and 4F. The comparison between learning and SGDrecurrent implementation is shown in Figure 4G.…”
Section: Recurrent and Learning-based Snnmentioning
confidence: 96%
“…The simulated robotic arm model has six DoF, out of which, here, we controlled five (the sixth DoF accounts for gripping). The model follows the physical arm design described in Zaidel et al 11 and is visualized in Figure 1. The arm features six links l 1À6 (12-, 30-, 6-, 20-, 10-, and 20-cm long, respectively) and five joints q 1À5 (rotated around the y, x, x, z, x axes, respectively).…”
Section: Robotic Modelmentioning
confidence: 99%
“…NEF is extensively used to design neuromorphic systems capable of visual perception 10 and motor control. 11 It serves as the foundation for Nengo, a Python-based "neural compiler," which translates high-level descriptions to low-level neural models. 12 NEF-inspired neuromorphic hardware designs 13 have been implemented in both analog and digital circuitry.…”
Summary
Inverse kinematics is fundamental for computational motion planning. It is used to derive an appropriate state in a robot's configuration space, given a target position in task space. In this work, we investigate the performance of fully connected and residual artificial neural networks as well as recurrent, learning-based, and deep spiking neural networks for conventional and geometrically constrained inverse kinematics. We show that while highly parameterized data-driven neural networks with tens to hundreds of thousands of parameters exhibit sub-ms inference time and sub-mm accuracy, learning-based spiking architectures can provide reasonably good results with merely a few thousand neurons. Moreover, we show that spiking neural networks can perform well in geometrically constrained task space, even when configured to an energy-conserved spiking rate, demonstrating their robustness. Neural networks were evaluated on NVIDIA's Xavier and Intel's neuromorphic Loihi chip.
“…Different approaches to design of expanded conventional controller's structures in [4][5][6][7][8][9] described. In compar-Distributed under creative commons license 4.0 ison with the much simpler PI control, which still attracts attention of the contemporary research, the design is yet more complicated also due to the fact that an increased speed of transients exhibits all modeling and tuning imperfections.…”
Section: Introduction: Self-organized Smart Control In Advanced Intelligent Roboticsmentioning
The technology of knowledge base remote design of the smart fuzzy controllers with the application of the "Soft / quantum computing optimizer" toolkit software developed. The possibility of the transmission and communication the knowledge base using remote connection to the control object considered. Transmission and communication of the fuzzy controller’s knowledge bases implemented through the remote connection with the control object in the online mode apply the Bluetooth or WiFi technologies. Remote transmission of knowledge bases allows designing many different built-in intelligent controllers to implement a variety of control strategies under conditions of uncertainty and risk. As examples, two different models of robots described (mobile manipulator and (“cart-pole” system) inverted pendulum). A comparison of the control quality between fuzzy controllers and quantum fuzzy controller in various control modes is presented. The ability to connect and work with a physical model of control object without using than mathematical model demonstrated. The implemented technology of knowledge base design sharing in a swarm of intelligent robots with quantum controllers. It allows to achieve the goal of control and to gain additional knowledge by creating a new quantum hidden information source based on the synergetic effect of combining knowledge. Development and implementation of intelligent robust controller’s prototype for the intelligent quantum control system of mega-science project NICA (at the first stage for the cooling system of superconducted magnets) is discussed. The results of the experiments demonstrate the possibility of the ensured achievement of the control goal of a group of robots using soft / quantum computing technologies in the design of knowledge bases of smart fuzzy controllers in quantum intelligent control systems. The developed software toolkit allows to design and setup complex ill-defined and weakly formalized technical systems on line.
Muscles are stretched with bursts of spikes that come from motor neurons connected to the cerebellum through the spinal cord. Then, alpha motor neurons directly innervate the muscles to complete the motor command coming from upper biological structures. Nevertheless, classical robotic systems usually require complex computational capabilities and relative high-power consumption to process their control algorithm, which requires information from the robot’s proprioceptive sensors. The way in which the information is encoded and transmitted is an important difference between biological systems and robotic machines. Neuromorphic engineering mimics these behaviors found in biology into engineering solutions to produce more efficient systems and for a better understanding of neural systems. This paper presents the application of a Spike-based Proportional-Integral-Derivative controller to a 6-DoF Scorbot ER-VII robotic arm, feeding the motors with Pulse-Frequency-Modulation instead of Pulse-Width-Modulation, mimicking the way in which motor neurons act over muscles. The presented frameworks allow the robot to be commanded and monitored locally or remotely from both a Python software running on a computer or from a spike-based neuromorphic hardware. Multi-FPGA and single-PSoC solutions are compared. These frameworks are intended for experimental use of the neuromorphic community as a testbed platform and for dataset recording for machine learning purposes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
