KIcker: An Industrial Drive and Control Foosball System automated with Deep Reinforcement Learning

This chapter discusses basic concepts from control theory and machine learning to facilitate a formal understanding of animal learning and motor control. It first distinguishes between feedback and feed-forward control strategies, and later introduces the classification of machine learning applications into supervised, unsupervised, and reinforcement learning problems. Next, it links these concepts with their counterparts in the domain of the psychology of animal learning, highlighting the analogies between supervised learning and classical conditioning, reinforcement learning and operant conditioning, and between unsupervised and perceptual learning. Additionally, it interprets innate and acquired actions from the standpoint of feedback vs anticipatory and adaptive control. Finally, it argues how this framework of translating knowledge between formal and biological disciplines can serve us to not only structure and advance our understanding of brain function but also enrich engineering solutions at the level of robot learning and control with insights coming from biology.

Download Full-text

Integrated process-system modelling and control through graph neural network and reinforcement learning

CIRP Annals ◽

10.1016/j.cirp.2021.04.056 ◽

2021 ◽

Author(s):

Jing Huang ◽

Jianjing Zhang ◽

Qing Chang ◽

Robert X. Gao

Keyword(s):

Neural Network ◽

Reinforcement Learning ◽

System Modelling ◽

Integrated Process ◽

Process System ◽

And Control

Download Full-text

Drone Deep Reinforcement Learning: A Review

Electronics ◽

10.3390/electronics10090999 ◽

2021 ◽

Vol 10 (9) ◽

pp. 999

Author(s):

Ahmad Taher Azar ◽

Anis Koubaa ◽

Nada Ali Mohamed ◽

Habiba A. Ibrahim ◽

Zahra Fathy Ibrahim ◽

...

Keyword(s):

Reinforcement Learning ◽

State Of The Art ◽

Real Life ◽

Environment Monitoring ◽

Simulated Environments ◽

Infrastructure Inspection ◽

Remote Sensing Mapping ◽

And Control ◽

The Military ◽

Uav Navigation

Unmanned Aerial Vehicles (UAVs) are increasingly being used in many challenging and diversified applications. These applications belong to the civilian and the military fields. To name a few; infrastructure inspection, traffic patrolling, remote sensing, mapping, surveillance, rescuing humans and animals, environment monitoring, and Intelligence, Surveillance, Target Acquisition, and Reconnaissance (ISTAR) operations. However, the use of UAVs in these applications needs a substantial level of autonomy. In other words, UAVs should have the ability to accomplish planned missions in unexpected situations without requiring human intervention. To ensure this level of autonomy, many artificial intelligence algorithms were designed. These algorithms targeted the guidance, navigation, and control (GNC) of UAVs. In this paper, we described the state of the art of one subset of these algorithms: the deep reinforcement learning (DRL) techniques. We made a detailed description of them, and we deduced the current limitations in this area. We noted that most of these DRL methods were designed to ensure stable and smooth UAV navigation by training computer-simulated environments. We realized that further research efforts are needed to address the challenges that restrain their deployment in real-life scenarios.

Download Full-text

Modeling human-like longitudinal driver model for intelligent vehicles based on reinforcement learning

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering ◽

10.1177/0954407020983579 ◽

2021 ◽

pp. 095440702098357

Author(s):

Ju Xie ◽

Xing Xu ◽

Feng Wang ◽

Haobin Jiang

Keyword(s):

Reinforcement Learning ◽

Comprehensive Evaluation ◽

Path Following ◽

Intelligent Vehicles ◽

Driver Model ◽

Control Center ◽

Training Performance ◽

Learning Agents ◽

System A ◽

And Control

The driver model is the decision-making and control center of intelligent vehicle. In order to improve the adaptability of intelligent vehicles under complex driving conditions, and simulate the manipulation characteristics of the skilled driver under the driver-vehicle-road closed-loop system, a kind of human-like longitudinal driver model for intelligent vehicles based on reinforcement learning is proposed. This paper builds the lateral driver model for intelligent vehicles based on optimal preview control theory. Then, the control correction link of longitudinal driver model is established to calculate the throttle opening or brake pedal travel for the desired longitudinal acceleration. Moreover, the reinforcement learning agents for longitudinal driver model is parallel trained by comprehensive evaluation index and skilled driver data. Lastly, training performance and scenarios verification between the simulation experiment and the real car test are performed to verify the effectiveness of the reinforcement learning based longitudinal driver model. The results show that the proposed human-like longitudinal driver model based on reinforcement learning can help intelligent vehicles effectively imitate the speed control behavior of the skilled driver in various path-following scenarios.

Download Full-text

A Novel Driving Scheme for Three-Phase Bearingless Induction Machine with Split Winding

Energies ◽

10.3390/en14164930 ◽

2021 ◽

Vol 14 (16) ◽

pp. 4930

Author(s):

Francisco Elvis Carvalho Souza ◽

Werbet Silva ◽

Andrés Ortiz Salazar ◽

José Paiva ◽

Diego Moura ◽

...

Keyword(s):

Control System ◽

Digital Signal Processor ◽

Position Control ◽

Radial Position ◽

Three Phase ◽

Control Configuration ◽

Two Phases ◽

Split Winding ◽

Drive And Control ◽

And Control

In order to reduce the costs of implementing the radial position control system of a three-phase bearingless machine with split winding, this article proposes a driving method that uses only two phases of the system instead of the three-phase traditional one. It reduces from six to four the number of inverter legs, drivers, sensors, and current controllers necessary to drive and control the system. To justify the proposal, this new power and control configuration was applied to a 250 W machine controlled by a digital signal processor (DSP). The results obtained demonstrated that it is possible to carry out the radial position control through two phases, without loss of performance in relation to the conventional three-phase drive and control system.

Download Full-text

Daily and Non-daily Smokers: A Profile of Drive and Cognitive Control Mechanisms

10.31237/osf.io/h9gpe ◽

2021 ◽

Author(s):

James Edward Bartlett

Keyword(s):

Systematic Review ◽

Attentional Bias ◽

Inhibitory Control ◽

Flanker Task ◽

Error Processing ◽

Future Research ◽

Meaningful Difference ◽

Addictive Behaviour ◽

Drive And Control ◽

And Control

Historically, smokers were considered a single homogeneous group, but over the past two decades research has increasingly focused on differentiating daily and non-daily smokers. Despite fundamentally different smoking habits and motives, daily and non-daily smokers have similar cessation rates. In order to understand why both groups may experience a similar difficulty quitting smoking, this thesis explored neurocognitive mechanisms associated with addictive behaviour. In order to profile these mechanisms, a systematic review was conducted, highlighting there was a gap to address in two areas of research relating to drive and control. Study One (N = 60) and Study Two (N = 166) investigated attentional bias towards smoking cues using the visual probe task, finding there was no meaningful difference between daily and non-daily smokers in trait-level attentional bias. Study Three (N = 28) measured ERP components associated with inhibitory control (Go/NoGo task) and error processing (Eriksen Flanker task). There were no significant effects of interest, but the sample size was smaller than planned. This thesis made three contributions to the study of addictive behaviour. First, the systematic review highlighted that research investigating lighter and heavier smokers has a problematic level of heterogeneity in the definitions used to define the groups. Second, there was no meaningful difference in attentional bias between daily and non-daily smokers, supporting contemporary theories that attentional bias may be best conceptualised as a state-level construct. Finally, internal consistency estimates of the ERP measures of inhibitory control and error processing supported previous research reporting good psychometric properties. Overall, this thesis presented a focused profile of measures relating to drive and control neurocognitive mechanisms, but there were no meaningful differences between daily and non-daily smokers. If these mechanisms are important to addictive behaviour, future research will have to investigate their role using alternative designs.

Download Full-text

Research on Real Working Condition Simulation and Performance Test of Wind Power Main Bearing Based on Test Bench

Mathematical Problems in Engineering ◽

10.1155/2021/6623988 ◽

2021 ◽

Vol 2021 ◽

pp. 1-13

Author(s):

Guodong Zhai ◽

Xujie Qin ◽

Xing Yang

Keyword(s):

Wind Power ◽

Performance Test ◽

Test Bench ◽

Test Results ◽

Main Bearing ◽

Loading System ◽

Modeling Software ◽

And Performance ◽

Drive And Control ◽

And Control

As a renewable energy source, wind energy has received more and more attention, and the wind power industry has also been advocated and developed by countries all over the world. In the production and use of wind turbines, the design and manufacturing technology of wind turbine bearings is very important. In order to ensure the reliable operation of the wind power main bearing after installation and realize the longest life of it, this paper designs a bearing test bench that can test the performance of the wind power main bearing. It can analyze the temperature, displacement, load, and moment of the key parts of the 5 MW wind power main shaft bearing. The solid modeling of the experimental platform was carried out using the 3D modeling software SolidWorks. Hydraulic loading system and test monitoring system are designed to realize the drive and control of the test bench. Through the established mathematical model, the central load of the hub is converted into the axial cylinder load and the radial cylinder load of the test bench to simulate the actual working conditions of the tested bearing. The test results show that the test bench meets various loading requirements and can reliably complete the task of testing wind power main bearings.

Download Full-text

The Use of Vibration as a Means of Industrial Drive

Proceedings of the Institution of Mechanical Engineers ◽

10.1243/pime_proc_1947_157_008_02 ◽

1947 ◽

Vol 157 (1) ◽

pp. 20-31

Author(s):

C. A. M. Thornton

Keyword(s):

Natural Frequency ◽

Transverse Vibration ◽

Vibration Amplitude ◽

Electrical Excitation ◽

Advantages And Disadvantages ◽

Transverse Vibrations ◽

Industrial Drive ◽

The Subject ◽

And Control

The application of vibration as a means of industrial drive has been considerably developed in the last twenty years. Starting from the obvious application of screening, it has been extended to conveying, heat interchanging, consolidation of material in packages to reduce shipping space, keeping material “fluid” in hoppers and chutes, hammering, etc. The paper discusses that part of the subject involved in the production and control of the vibration by mechanical or electrical means under all conditions of load. Formulae are developed for calculation of spring strength and for the calculation of the spring dimensions, and for the avoidance of spring fatigue. The various methods of electrical excitation are compared, and the relative advantages and disadvantages are discussed. The desirable instrument equipment is described, including the remote indication of vibration amplitude. A method is outlined by which it is claimed that vibration can be maintained constant at all loads and at any desired frequency. The testing of vibrating drives at the manufacturer's works and on site is discussed. In an Appendix to the paper the problem of transverse vibrations in long vibrating conveyors is treated mathematically, and a formula is developed for the natural frequency of transverse vibration of a conveyor of any uniform section and of any length.

Download Full-text