Command Input Research Articles

Variable task switching in the feeding network of Aplysia is a function of differential command input.

Command systems integrate sensory information and then activate the interneurons and motor neurons that mediate behavior. Much research has established that the higher-order projection neurons that constitute these systems can play a key role in specifying the nature of the motor activity induced, or determining its parametric features. To a large extent, these insights have been obtained by contrasting activity induced by stimulating one neuron (or set of neurons) to activity induced by stimulating a different neuron (or set of neurons). The focus of our work differs. We study one type of motor program, ingestive feeding in the mollusc Aplysia californica, which can either be triggered when a single projection neuron (CBI-2) is repeatedly stimulated or can be triggered by projection neuron coactivation (e.g., activation of CBI-2 and CBI-3). We ask why this might be an advantageous arrangement. The cellular/molecular mechanisms that configure motor activity are different in the two situations because the released neurotransmitters differ. We focus on an important consequence of this arrangement, the fact that a persistent state can be induced with repeated CBI-2 stimulation that is not necessarily induced by CBI-2/3 coactivation. We show that this difference can have consequences for the ability of the system to switch from one type of activity to another.NEW & NOTEWORTHY We study a type of motor program that can be induced either by stimulating a higher-order projection neuron that induces a persistent state, or by coactivating projection neurons that configure activity but do not produce a state change. We show that when an activity is configured without a state change, it is possible to immediately return to an intermediate state that subsequently can be converted to any type of motor program.

Effects of Sitting Recovery Protocols on Postexercise Heart Rate in Young Adults

Abstract Background We aimed to compare the effects of 3 different upright seated recovery protocols on heart rate recovery (HRR) after a submaximal ergometer exercise. Methods The study included 30 young adult volunteers (15 men and 15 women). Participants performed a submaximal cycle exercise test at a constant workload of 60 Watts until a steady HR was achieved. This was followed by 5 min of: (a) inactive, (b) active loadless, and (c) passive recovery protocols. The HRR was assessed as the difference between the peak exercise HR and the HR recorded following 1 min of recovery and as the percentage HR decline after 1 min postexercise. Abnormal HRR was defined as a reduction of 1-min HRR ≤12 b · min−1 or ≤15 b · min−1 or ≤18 b · min−1. Results In both sexes, HRR was faster during inactive recovery compared with active recovery. In males, HRR was faster during inactive recovery compared with passive recovery protocol. In females, HRR was faster during passive recovery compared with active recovery protocol. The prevalence of impaired HRR was greater in the active recovery compared with the passive and inactive recovery protocols in both sexes. Conclusion The present findings suggest that in both sexes, the postexercise HRR was mediated by a combined action of the central command and other inputs or stimuli arising from skeletal muscle activities. In addition, the active recovery protocol resulted to a slower HRR and elicited more abnormal postexercise HR responses compared with the other recovery protocols.

Pilot Command Implementation in Aircrafts: A New Approach Based on Dynamic Inversion

A new approach based on dynamic inversion is proposed in this paper for implementing pilot commands in an aircrafts. The command inputs from the pilot are assumed to be (i) the normal acceleration and forward velocity commands in the longitudinal mode and (ii) roll rate, height and forward velocity commands in the lateral mode. A major difference here is that the second derivatives of the velocities along body y and z directions are assumed to be zero, as opposed to the first derivatives (which is used in many published literature). The new approach leads to a significant reduction of tuning parameters in the control design process, which is a major advantage. Detailed derivations of the modified approach are presented with respect to the generic aircraft dynamics model available in [16]. Extensive Six-DOF simulation studies show that, besides the above advantage, the new approach leads to two other additional advantages; namely reduced oscillatory response and reduced control magnitude. In a comparison study with an existing method using the Six-DOF model for Boeing 747 (with the numerical data available in [16]), numerical results clearly show the improved performance of this new approach. Furthermore, in lateral mode a technique for obtaining a corresponding roll rate command from the desired bank angle command is also implemented. Comparison results show improved performance of the new approach with this modification as well.

Overview of System Identification Tool Advancements at Systems Technology, Incorporated

For nearly two decades, Systems Technology, Inc. has been investigating novel system identification methods and practical applications. Two such methods will be highlighted herein: 1) the use of wavelet-based scalograms to explore the nature of dynamic systems in both the time and frequency domains simultaneously; and 2) a method to estimate frequency, damping, and mode shapes for lightly damped aeroelastic modes exhibited across a variety of air vehicles. From past work, two scalogram-based metrics have evolved, which are the power frequency and inceptor peak power phase, that will be one focus of this paper. The power frequency provides a metric that considers not only the dominant frequencies of the signal of interest, such as the command input from a pilot, but also the power at those frequencies. For the inceptor peak power phase, scalograms are used to identify the peak input power of a pilot’s inceptor command signal that when compared against a pitch or roll rate phase angle at the peak frequency produces a new method to detect the onset of pilot-induced oscillations. This paper will further describe system identification tools that extract modal information of aeroelastic vibrations observed in flight. Because it is often difficult to obtain input excitation data, particularly in real time, Systems Technology, Inc. developed a modal identification software toolbox that provides an innovative approach based on curve-fitting frequency domain decomposition, which relies only on output sensor data to identify the aeroelastic modes. Flight-test data are used to demonstrate the effectiveness and efficacy of the system identification approaches.

Experimental Application of Global Fast Terminal Sliding Mode Controller to TITO System

Comprehensive analysis of a global fast terminal sliding mode control strategy (GFTSMC) for the multivariable laboratory level control system is presented in this paper. The performance of conventional sliding mode controller strategies experiences chattering as switching control input contains the sgn’ function and singularity issues. Moreover, the performance is degraded due to parametric uncertainties and external disturbances. Robustness issues are not well defined in conventional strategies. Global fast terminals remove the chattering effect and eliminate the singularity problem. It has a shorter convergence time and better reaching precision. It shows the finite-time convergence of output variables to the command input. Indirect stability is guaranteed using the direct Lyapunov function. To elicit the performance of the proposed strategy, simulation tests have been conducted on the wood-berry distillation process, while the experimental tests are carried out on a laboratory multivariable process control system. The performance of the proposed strategy is compared to that of conventional sliding mode control and fast terminal sliding mode control.

Goal-Conditioned Generators of Deep Policies

Goal-conditioned Reinforcement Learning (RL) aims at learning optimal policies, given goals encoded in special command inputs. Here we study goal-conditioned neural nets (NNs) that learn to generate deep NN policies in form of context-specific weight matrices, similar to Fast Weight Programmers and other methods from the 1990s. Using context commands of the form ``generate a policy that achieves a desired expected return,'' our NN generators combine powerful exploration of parameter space with generalization across commands to iteratively find better and better policies. A form of weight-sharing HyperNetworks and policy embeddings scales our method to generate deep NNs. Experiments show how a single learned policy generator can produce policies that achieve any return seen during training. Finally, we evaluate our algorithm on a set of continuous control tasks where it exhibits competitive performance. Our code is public.

Rancang Bangun Sistem Smart Door Lock Menggunakan Mikrokontroler Esp32 Berbasis Internet Of Things (Iot) dan Smartphone Android

This study aims to design and build smart door lock system using Microcontroller ESP32 based on Internet of Things (IoT) and Smartphone Android which will be installed in one of the study rooms at Pahlawan Tuanku Tambusai University using the Quality Function Deployment (QFD) method which aims to identify and meet the needs and desires of consumers for the resulting product or service. This smart door lock is specially made to improve the security system more optimally, and transforming the use from manual door locks to smart door locks that more ergonomic in following technological developments and advances and expanding the use of internet networks in the field of security systems. The process of manufacture smart door locks based on consumer desires using the QFD method which has been processed using the house of quality (HoQ). it can be seen that the percentage of the highest value or priority for product attributes is “Control the smart door lock system using a smartphone” with a value presentation of 5,643%, followed by “Access the smart door lock using the internet network"with a value of 5,479%, Then on the third priority, “Smart door lock system accurately accepts input commands” with a value of 5,226%. Smart door lock product specifications based on the voice of customer statement are that in the main function of the smart door lock system can be controlled using a smartphone with blynk application control software via the internet network, then to meet the wishes of consumers provided wifi module on system components, or the durability of the door lock lever (solenoid) made of steel coated with stenliss so that it is not easily break in, as well the aesthetics (design) of smart door locks using PVC Box as a hardware display, and equipped with alarm as indication of sound when the smart door lock system is successfully accessed.

Research on the Control Method of Micro-gas Turbine Power Supply Modulation System for Primary Energy in Electromagnetic Launch Technology

Combining the advantages of micro-gas turbine generation system (MTGS) and power converters, moreover, comparing the various design ideas for the primary energy of electromagnetic launch technology, we design a primary energy to meet the reliability, stability, safety and portable of electromagnetic launch technology. We discuss the control ideas and methods to be adopted in this system, the first step is to output a stable DC bus voltage through rectification, and then to step up the voltage through a series resonant converter to achieve the desired output voltage. Through theoretical analysis and simulation, it reveals the effect of this load on the performance of MTGS, and the parameter optimization design is combined with the voltage outer loop control strategy to achieve the purpose of outputting stable voltage. We used a ARM controller to implement the control of this system to measure the response curve of the system under different input commands. According to the theoretical analysis, numerical simulation and experimental results, the control strategy designed in this paper can better meet the performance index requirements of micro-gas turbine engine (MTE) and power converter as well as the control method is simple and easy to implement in engineering. The results of this paper confirm the feasibility of this primary energy application for electromagnetic launch technology.

Roadway construction noise model progression and integration with federal highway administration's traffic noise model

In 2006, the Federal Highway Administration (FHWA) published the Roadway Construction Noise Model (RCNM 1.0) to predict noise levels from construction activities. RCNM 1.0 included basic acoustical data for a variety of construction equipment. In 2018, RCNM 2.0 was developed using a similar user interface to FHWA's Traffic Noise Model (TNM) version 3.0. RCNM 2.0 improved prediction of construction noise propagation and mitigation. RCNM 2.0 also expanded the source database and allowed users to analyze the project site conditions using mapping. TNM 3.2 includes integration of the TNM 3.1 software structure with RCNM 2.0. Equipment source data in the RCNM database were calibrated to account for the excess attenuation from each new source height. The RCNM database consists of moving and stationary equipment sources, each with spectra for two-four cardinal directions relative to the equipment, which may not always operate constantly throughout the day. Modeling these novel source features in TNM required new logic to handle stationary point sources, partially inactive sources, and source directionality. The goal of TNM 3.2 was for analysts to combine roadway and construction noise analyses within one file. Doing so required changes to the TNM object intersection rules, as well as all aspects of the user interface including map displays, input commands, and output data tables.

Dynamics and Control of a Buoyancy-Driven Underwater Vehicle for Estimating and Tracking the Scattering Layer

This paper presents the design of a control strategy for an autonomous buoyancy-driven underwater vehicle, called the Driftcam, tasked with estimating and tracking the pelagic acoustic-scattering layer at depths of up to 700 meters. This reference-tracking control design is an alternative to the onboard control currently implemented on the Driftcam, as the existing control architecture is not designed for time-varying setpoint tracking. Tracking a known reference is presented using state-feedback control for continuous- and discrete-time command inputs. The known reference signal is chosen to represent the pelagic scattering layer vertical migration dynamics, modeled as a simple harmonic oscillator and estimated with a dynamic observer. We define a measurement function for the depth-varying density of organisms observable in the scattering layer. Time-varying reference tracking control is presented using output feedback for continuous- and discrete-time command inputs, where the reference is the online position estimate of the scattering layer. We also present an offboard trajectory motion planning algorithm using the existing onboard controller to emulate the response of the proposed reference tracking strategy. Experimental results from laboratory and field testing of the trajectory motion planner demonstrate that the buoyancy engine can feasibly achieve the desired behavior of the reference-tracking control strategy.

Vehicle Control on an Uninstrumented Surface With an Off-the-Shelf Smartwatch

Advances in wearable and machine learning technologies have caused smartwatches to emerge as promising input devices across disciplines. In this study, we propose a user interface, designed for an automotive environment for recognizing input commands. As the use of extra elements to extend the scope of interactions may visually distract users during driving, we focus on using body parts as an interaction space. Specifically, we utilized the lap as an interaction hyperplane due to its near-flat surface, which provides users interaction analogous to a touchpad. To substantiate the proposed approach, we collected motion signals using an off-the-shelf smartwatch and trained them in supervised learning settings. As target gestures, we defined ten different hand gestures, such as tapping and folding/spreading all fingers on the lap. The experimental results exhibited a test accuracy of 94%, thereby validating the feasibility of the proposed approach.

Ability of ChatGPT to generate competent radiology reports for distal radius fracture by use of RSNA template items and integrated AO classifier

The amount of acquired radiology imaging studies grows worldwide at a rapid pace. Novel information technology tools for radiologists promise an increase of reporting quality and as well quantity at the same time. Automated text report drafting is one branch of this development. We defined for the present study in total 9 cases of distal radius fracture. Command files structured according to a template of the Radiological Society of North America (RSNA) and to Arbeitsgemeinschaft Osteosynthese (AO) classifiers were given as input to the natural language processing tool ChatGPT. ChatGPT was tasked with drafting an appropriate radiology report. A parameter study (n=5 iterations) was performed. An overall high appraisal of ChatGPT radiology report quality was obtained in a score card based assessment. ChatGPT demonstrates the capability to adjust output files in response to minor changes in input command files. Existing shortcomings were found in technical terminology and medical interpretation of findings. Text drafting tools might well support work of radiologists in the future. They would allow a radiologist to focus time on the observation of image details and patient pathology. ChatGPT can be considered a substantial step forward towards that aim.

Unifying Control Architecture for Reactive Particle Swarms

Swarm control strategies allow for decentralized control of many simple robots to perform collective behaviors based on local interactions. Despite the broad applications of swarm robotics, there lacks a generalized and unified set of control equations for swarms. First, this work presents a novel mathematical control law for decentralized velocity control of planar, homogeneous swarms called reactive particle swarms (RPS). The architecture unifies the development of new RPS behaviors. The weighted summation of simple base behaviors and external command inputs form complex composite behaviors. Second, a series of simulated and on-hardware case studies demonstrate the utility and flexibility of the architecture. Third, this work establishes a library of verified RPS base behaviors.

Forearm Dual-Triboelectric Sensor (FDTS) for assistive Human-Machine-Interfaces (HMIs) and robotic control with potential uses in prosthetic devices

This work describes a new sensitive non-intrusive forearm device with two triboelectric sensors (TS) integrated into one flexible PCB (FPCB) that indirectly differentiates finger and hand movements based on the forearm muscles and tendons' combined responses. The new Forearm Dual Triboelectric Sensor (FDTS) intends to demonstrate triboelectric nanogenerators (TENGs) as an affordable and flexible alternative to expensive and bulky EMG devices used in prosthetic limbs by amputated patients. The proposed concept can successfully recognize up to 10 different input signals from the bending or tapping of the middle and little fingers and hand opening-closing action used as input commands for a Human-Computer-Interface (HCI) for computer navigation. In the experiments, we can successfully control horizontal, vertical and diagonal movements, left and right click of the cursor on the screen, and scroll up/down options. The FDTS was also implemented in robotic hand remote control, where different index and small finger bending amplitudes were translated to different mechanical finger rotation degrees and overall robot hand movement. Additionally, the robotic hand was able to perform right and left clicks controlled by fast finger bending. More experiments demonstrate the FDTS response to other wrist movements, such as pronation and supination. This proves the potential benefits of TS in cost-effective prosthetic devices, with limited but better functionality than standard body-powered prostheses. In the future, FDTS-like devices can be modularly integrated with artificial limbs to allow amputated patients to control computers and external devices. Additionally, this solution can positively impact the integration of online spaces like the Metaverse with healthcare technologies for rehabilitation and social inclusion.

Reference Model-Based Deterministic Policy for Pitch and Depth Control of Autonomous Underwater Vehicle

The Deep Reinforcement Learning (DRL) algorithm is an optimal control method with generalization capacity for complex nonlinear coupled systems. However, the DRL agent maintains control command saturation and response overshoot to achieve the fastest response. In this study, a reference model-based DRL control strategy termed Model-Reference Twin Delayed Deep Deterministic (MR-TD3) was proposed for controlling the pitch attitude and depth of an autonomous underwater vehicle (AUV) system. First, a reference model based on an actual AUV system was introduced to an actor–critic structure, where the input of the model was the reference target, the outputs were the smoothed reference targets, and the reference model parameters can adjust the response time and the smoothness. The input commands were limited to the saturation range. Then, the model state, the real state and the reference target were mapped to the control command through the Twin Delayed Deep Deterministic (TD3) agent for training. Finally, the trained neural network was applied to the AUV system environment for pitch and depth experiments. The results demonstrated that the controller can eliminate the response overshoot and control command saturation while improving the robustness, and the method also can extend to other control platforms such as autonomous guided vehicle or unmanned aerial vehicle.

Fast Obstacle Avoidance Based on Real-Time Sensing

Humans excel at navigating and moving through dynamic and complex spaces, such as crowded streets. For robots to do the same, it is crucial that they are endowed with highly reactive obstacle avoidance which is adept at partial and poor sensing. We address the issue of enabling obstacle avoidance based on sparse and asynchronous perception. The proposed control scheme combines a high-level input command provided by either a planner or a human operator with fast reactive obstacle avoidance (FOA). The sampling-based sensor data can be combined with an analytical reconstruction of the obstacles for real-time collision avoidance. Thus, we can ensure that the agent does not become stuck when a feasible path exists between obstacles. Our algorithm was evaluated experimentally on static laser data from cluttered, indoor office environments. Additionally, it was used in shared-control mode in a dynamic and complex outdoor environment in the center of Lausanne. The proposed control scheme successfully avoided collisions in both scenarios. During the experiments, the controller took 1 ms to evaluate over 30000 data points.

Multimodal bipedal locomotion generation with passive dynamics via deep reinforcement learning.

Generating multimodal locomotion in underactuated bipedal robots requires control solutions that can facilitate motion patterns for drastically different dynamical modes, which is an extremely challenging problem in locomotion-learning tasks. Also, in such multimodal locomotion, utilizing body morphology is important because it leads to energy-efficient locomotion. This study provides a framework that reproduces multimodal bipedal locomotion using passive dynamics through deep reinforcement learning (DRL). An underactuated bipedal model was developed based on a passive walker, and a controller was designed using DRL. By carefully planning the weight parameter settings of the DRL reward function during the learning process based on a curriculum learning method, the bipedal model successfully learned to walk, run, and perform gait transitions by adjusting only one command input. These results indicate that DRL can be applied to generate various gaits with the effective use of passive dynamics.

Optimizing Hand Gesture Recognition Using CNN Model Supported by Raspberry pi for Self-Service Technology

This study describes the optimization of hand gesture recognition on Raspberry Pi 4 technology has advanced over the past years, some computers are now able to compute much more complex problems like real-time object detection. But for small devices, optimization is required to run in real-time with acceptable performance in terms of latency and low-cost effect on accuracy. Low latency is a requirement for most technology, especially when integrating real-time object detection as input into Self-Service Technology on Raspberry Pi for the store. This research was conducted on 288 pictures with six types of chosen hand gestures for command inputs that have been configured in the Self-Service Technology as a training dataset. In the experiment carried out with 5 CNN object detection models were used, namely YOLOv3-Tiny-PRN, YOLOv4-Tiny, MobileNetV2-Yolov3-NANO, YOLO-Fastest-1.1, and YOLO-Fastest-1.1-XL. Based on the experiment after optimization, the FPS and inference time metrics have improved performance. The performance improves due to a gained average value of FPS by 3 FPS and a reduced average value of inference time by 119,260 ms. But such an improvement also comes with a reduction in overall accuracy. The rest of the parameters have a reduced score on Precision, Recall, F1-Score, and some for IoU. Only YOLO-Fastest-1.1-XL have an improved value of IoU by about 0.58%. Some improvements in the CNN and dataset might improve the performance even more without sacrificing too much on the accuracy, but it's most likely suitable for another research as a continuation of this topic.

DESIGN AND IMPLEMENTATION OF A 36-NOTE LOW-COST CUSTOMIZED DIGITAL MUSIC BOX

The purpose of the current study was to design and build a music box with digitally produced notes. The input command sheet allows anybody to create any melody from the octave, making it possible to think of it as a small version of a musical instrument. This instrument is capable of producing frequencies in all three octaves, i.e., the lower, the middle, and the higher octaves. This music box may therefore play any tune, no matter how short or lengthy the note is. There is no need to alter the frequencies over time because no string is used in the creation. It can play any song to one’s heart’s content if the song’s music sheet is accessible. In an epitome, this music box can be viewed as a very simple musical instrument with digitally created tones. There is no need to customize the strings over time like with traditional mechanical musical instruments owing to tones produced digitally. The ability to generate a vast variety of tones, the ease with which it can be adjusted, the convenience with which anyone may have their desired notes customized, and the low cost at which it was produced are the most notable advantages of this digital music instrument. Due to its low price, it can be simply incorporated, altered, slightly improved, and produced at a considerably reduced cost for commercial use.

DESIGN AND IMPLEMENTATION OF A 36-NOTE LOW-COST CUSTOMIZED DIGITAL MUSIC BOX

The purpose of the current study was to design and build a music box with digitally produced notes. The input command sheet allows anybody to create any melody from the octave, making it possible to think of it as a small version of a musical instrument. This instrument is capable of producing frequencies in all three octaves, i.e., the lower, the middle, and the higher octaves. This music box may therefore play any tune, no matter how short or lengthy the note is. There is no need to alter the frequencies over time because no string is used in the creation. It can play any song to one’s heart’s content if the song’s music sheet is accessible. In an epitome, this music box can be viewed as a very simple musical instrument with digitally created tones. There is no need to customize the strings over time like with traditional mechanical musical instruments owing to tones produced digitally. The ability to generate a vast variety of tones, the ease with which it can be adjusted, the convenience with which anyone may have their desired notes customized, and the low cost at which it was produced are the most notable advantages of this digital music instrument. Due to its low price, it can be simply incorporated, altered, slightly improved, and produced at a considerably reduced cost for commercial use.