关键词： image-based sensor systems   object detection systems   safety of autonomous agriculture machines   test stand   humanoid test target  

摘要： As an essential part for the development of autonomous agricultural robotics, the functional safety of autonomous agricultural machines is largely based on the functionality and robustness of non-contact sensor systems for human protection. This article presents a new step in the development of autonomous agricultural machine with a concept and the realization of a novel test method using a dynamic test stand on an agricultural farm in outdoor areas. With this test method, commercially available sensor systems are tested in a long-term test around the clock for 365 days a year and 24 h a day on a dynamic test stand in continuous outdoor use. A test over a longer period of time is needed to test as much as possible all occurring environmental conditions. This test is determined by the naturally occurring environmental conditions. This fact corresponds to the reality of unpredictable/determinable environmental conditions in the field and makes the test method and test stand so unique. The focus of the developed test methods is on creating own real environment detection areas (REDAs) for each sensor system, which can be used to compare and evaluate the autonomous human detection of the sensor systems for the functional safety of autonomous agricultural robots with a humanoid test target. Sensor manufacturers from industry and the automotive sector provide their sensor systems to have their sensors tested in cooperation with the TuV.

关键词： autonomous navigation   collision-free   deep reinforcement learning   unmanned aerial vehicle  

摘要： Autonomous navigation and collision avoidance missions represent a significant challenge for robotics systems as they generally operate in dynamic environments that require a high level of autonomy and flexible decision-making capabilities. This challenge becomes more applicable in micro aerial vehicles (MAVs) due to their limited size and computational power. This paper presents a novel approach for enabling a micro aerial vehicle system equipped with a laser range finder to autonomously navigate among obstacles and achieve a user-specified goal location in a GPS-denied environment, without the need for mapping or path planning. The proposed system uses an actor-critic-based reinforcement learning technique to train the aerial robot in a Gazebo simulator to perform a point-goal navigation task by directly mapping the noisy MAV's state and laser scan measurements to continuous motion control. The obtained policy can perform collision-free flight in the real world while being trained entirely on a 3D simulator. Intensive simulations and real-time experiments were conducted and compared with a nonlinear model predictive control technique to show the generalization capabilities to new unseen environments, and robustness against localization noise. The obtained results demonstrate our system's effectiveness in flying safely and reaching the desired points by planning smooth forward linear velocity and heading rates.

关键词： Nanotechnology   Artificial intelligence   Automation   Personalised medicine   Microfluidics  

摘要： The field of nanotechnology and personalised medicine is undergoing drastic changes in the approach and efficiency of experimentation. The COVID-19 pandemic has spiralled into mass stagnation of major laboratories around the globe and led to increased investment into remote systems for nanoparticle experiments. A significant number of laboratories now operate using automated systems;however, the extension to nanoparticle preparation and artificial intelligence-dependent databases holds great translational promise. The strive to combine automation with artificial intelligence (AI) grants the ability to optimise targeted therapeutic nanoparticles for unique cell types and patients. In this perspective, the current and future trends of automated approaches to nanomedicine synthesis are discussed and compared with traditional methods.

关键词： Dynamic Neural Field   Neural integrator   Assembly robot   Value-based decision making   Sequence learning  

摘要： Modern manufacturing and assembly environments are characterized by a high variability in the built process which challenges human-robot cooperation. To reduce the cognitive workload of the operator, the robot should not only be able to learn from experience but also to plan and decide autonomously. Here, we present an approach based on Dynamic Neural Fields that apply brain-like computations to endow a robot with these cognitive functions. A neural integrator is used to model the gradual accumulation of sensory and other evidence as time-varying persistent activity of neural populations. The decision to act is modeled by a competitive dynamics between neural populations linked to different motor behaviors. They receive the persistent activation pattern of the integrators as input. In the first experiment, a robot learns rapidly by observation the sequential order of object transfers between an assistant and an operator to subsequently substitute the assistant in the joint task. The results show that the robot is able to proactively plan the series of handovers in the correct order. In the second experiment, a mobile robot searches at two different workbenches for a specific object to deliver it to an operator. The object may appear at the two locations in a certain time period with independent probabilities unknown to the robot. The trial-by-trial decision under uncertainty is biased by the accumulated evidence of past successes and choices. The choice behavior over a longer period reveals that the robot achieves a high search efficiency in stationary as well as dynamic environments.

关键词： Contact force estimation   Floating-base space robots   Nonlinear observer   Collision detection and isolation  

摘要： In this paper, the contact force detection, isolation and estimation problems for an orbital robot are addressed. First, an observer based on a base-joints dynamics is reviewed and analyzed considering the issues related to a space robotic application. Then, a new scheme is derived reformulating the dynamics in terms of the motion of the centroid of the whole robot and the joints. In both observers, three momentum-based residuals are defined, which can be used to reconstruct the external contact wrench. The reconstruction through the proposed observer turns out to provide superior performance with respect to the base-joints one thanks to the decoupling of the angular and joint momentum residuals from the spacecraft linear velocity, which is an inaccurate measurement in real space scenarios. The advantages of the proposed observer are shown through both simulative and hardware validation featuring a 7 degrees of freedom (DoF) arm mounted on a 6 DoF moving base.

关键词： Walking rehabilitation   Inverse dynamics   Power assist  

摘要： This paper describes a gait training algorithm based on the inverse dynamics of DDgo Pro, an end-effector-type walking rehabilitation robot. As a major feature, a gait training algorithm with high responsiveness and flexibility was achieved by analyzing the inverse dynamics of the five-link mechanism with a foothold attached to the end of the mechanism and compensating for the required torque for the link motion and the patient's weight during gait training. To this end, inverse dynamics was solved using Newton Euler's method, and three training modes were proposed. First, passive mode training performs three tasks: motion compensation of the five-link mechanism for the given constant walking speed through the inverse dynamics, patient's weight compensation, and tracking of the gait trajectory of the foothold using PD control to practice the normal walking pattern and restore muscle strength for standing. Second, Active-assisted mode training was implemented for the patient to receive muscle assistance from an electric motor in proportion to the patient's walking intention in addition to the reduced compensations of the motion of the five-link mechanism and patient's weight. Third, active mode training was implemented so that the patient can perform gait training while feeling their own weight with the time-varying motion compensation of the five-link mechanism. Finally, the performance of each gait training mode was experimentally verified, and the muscle activity was measured using EMG sensors to compare the muscle assistance performances of the training modes. Furthermore, the correction of the gait motion of the affected leg according to the motor assistance in active-assisted mode was also verified.

关键词： Programming   Sense-reason-act   Robotics   Computational thinking   Mathematics   Self-efficacy  

摘要： This study investigates the development of algorithmic thinking as a part of computational thinking skills and self-efficacy of primary school pupils using programmable robots in different instruction variants. Computational thinking is defined in the context of twenty-first century skills and describes processes involved in (re)formulating a problem in a way that a computer can process it. Programming robots offers specific affordances as it can be used to develop programs following a Sense-Reason-Act (SRA) cycle. The literature provides evidence that programming robots has the potential to enhance algorithmic thinking as a component of computational thinking. Specifically there are indications that pupils who use SRA-programming learn algorithmic skills better and achieve a higher level of self-efficacy in an open, scaffold learning environment than through direct instruction. In order to determine the influence of the instruction variant used, an experimental research design was made in which pupils solved algorithm-based mathematical problems (grid diagrams) in a preliminary measurement and their self-efficacy determined via a questionnaire. As an intervention, pupils learn to solve programming issues in pairs using "Lego NXT" robots and "Mindstorms" software in two instruction variants. The post-measurement consists of a Lego challenge, solving mathematical problems (grid diagrams), and a repeated self-efficacy questionnaire. This research shows an increase of our measures on algorithmic thinking dependent on the amount of SRA usage (though not significant). Programming using the SRA-cycle can be considered as the cause of the measured effect. The instruction variant used during the robotic intervention seems to play only a marginal role.

关键词： 3D printing   Sensor   Robotics   Piezoelectric  

摘要： Three-dimensional (3D) printing has become an important fabrication method for soft robotics, due to its ability to make complex 3D structures from computer designs in simple steps and multimaterial co-deposition ability. In this article, the application of 3D printing techniques in the fabrication of four types of tactile sensors commonly used in soft robotics, including the piezoresistive tactile sensor, capacitive tactile sensor, piezoelectric tactile sensor, and triboelectric tactile sensor, will be discussed. The 3D printing mechanism, material, and structure for each type of sensor will be introduced, and the perspectives on the future potential of 3D printable tactile sensors will be discussed.

关键词： Machine intelligence   Artificial intelligence   Neural networks  

摘要： Purpose For efficient trajectory control of industrial robots, a cumbersome computation for inverse kinematics and inverse dynamics is needed, which is usually developed using spatial transformation using Denavit-Hartenberg principle and Lagrangian or Newton-Euler methods, respectively. The model is highly non-linear and needs to deal with uncertainties because of lack of accurate measurement of mechanical parameters, noise and non-inclusion of joint friction, which results in some inaccuracies in predicting accurate torque trajectories. To get a guaranteed closed form solution, the robot designers normally follow Pieper's recommendation and compromise with the mechanical design. While this may be acceptable for the industrial robots where the aesthetic look is not that important, it is not for humanoid and social robots. To help solve this problem, this study aims to propose an alternative machine learning-based computational approach based on a multi-gated sequence model for finding appropriate mapping between Cartesian space to joint space and motion space to joint torque space. Design/methodology/approach First, the authors generate sufficient data required for the sequence model, using forward kinematics and forward dynamics by running N number of nested loops, where N is the number of joints of the robot. Subsequently, to develop a learning-based model based on sequence analysis, the authors propose to use long short-term memory (LSTM) and hence, train an LSTM model, the architecture details of which have been discussed in the paper. To make LSTM learning algorithms perform efficiently, the authors need to detect and eliminate redundant features from the data set, which the authors propose to do using an elegant statistical tool called Pearson coefficient. Findings To validate the proposed model, the authors have performed rigorous experiments using both hardware and simulation robots (Baxter/Anukul robot) available in their laboratory and KUKA simulation robo