Behavior Trees (BTs) were invented as a tool to enable modular AI in computer games, but have received an increasing amount of attention the robotics community last decade. With rising demands on agent complexity, game programmers found that Finite State Machines (FSM) they used scaled poorly and difficult extend, adapt reuse. In BTs, state transition logic is not dispersed across individual states, organized hierarchical tree structure, with states leaves. This has significant effect...

This paper presents a unified framework for Behavior Trees (BTs), plan representation and execution tool. The available literature lacks the consistency mathematical rigor required robotic control applications. Therefore, we approach this problem in two steps: first, reviewing most popular BT exposing aforementioned issues; second, describing our along with equivalence notions between BTs Controlled Hybrid Dynamical Systems (CHDSs). improves on existing state of art as it describes more...

Human beings subconsciously adapt their behaviors to a communication partner in order make interactions run smoothly. In human-robot interactions, not only the human but also robot is expected its partner. Thus, facilitate should be able read subconscious comfort and discomfort signals from humans adjust behavior accordingly, just like would. However, most previous research works consciously give feedback, which might interfere with aim of interaction. We propose an adaptation mechanism...

Today, microbial drinking water quality is monitored through either time-consuming laboratory methods or indirect on-line measurements. Results are thus delayed insufficient to support proactive action. A novel, optical, bacteria sensor with a 10-minute time resolution has been developed. The based on 3D image recognition, and the obtained pictures analyzed algorithms considering 59 quantified parameters. counts individual suspended particles classifies them as abiotic particles. technology...

In recent decades, there has been an increasing interest in the use of robotic powered exoskeletons to assist patients with movement disorders rehabilitation and daily life. Providing assistive torque that compensates for user's remaining muscle contributions is a growing challenging field within exoskeleton control. this article, ankle joint torques were estimated using electromyography (EMG)-driven neuromusculoskeletal (NMS) model artificial neural network (ANN) seven tasks, including fast...

Figure 1.A robot distributing a flyer to pedestrian A model of distributional handing interaction for mobile Abstract-This paper reports our research on developing flyers pedestrians.The difficulty is that potential receivers are pedestrians who not necessarily cooperative; thus, the needs appropriately plan its motion, making it easy and non-obstructive receive flyers.In order establish model, we observed human interactions in real world.We analyzed evaluated different methods people...

In this work we present an adaptive control approach for pivoting, which is in-hand manipulation maneuver that consists of rotating a grasped object to desired orientation relative the robot's hand. We perform pivoting by means gravity, allowing rotate between fingers one degree freedom gripper and controlling gripping force ensure follows reference trajectory arrives at angular position. use visual pose estimation system track measurements from tactile sensors force. The controller employs...

Industrial robots can solve tasks in controlled environments, but modern applications require able to operate also unpredictable surroundings. An increasingly popular reactive policy architecture robotics is Behavior Trees (BTs) as other architectures, programming time drives cost and limits flexibility. The two main branches of algorithms generate policies automatically, automated planning machine learning, both have their own drawbacks not previously been combined for generation BTs. We...

Gait phase recognition is of great importance in the development assistance-as-needed robotic devices, such as exoskeletons. In order for a powered exoskeleton with phase-based control to determine and provide proper assistance wearer during gait, user’s current gait must first be identified accurately. can potentially achieved through input from wearable sensors. Deep convolutional neural networks (DCNN) machine learning approach that widely used image recognition. User kinematics, measured...

In this work we propose a sliding mode controller for in-hand manipulation that repositions tool in the robot's hand by using gravity and controlling slippage of tool. our approach, robot holds with pinch grasp model system as link attached to gripper via passive revolute joint friction, i.e., only affords rotational motions around given axis rotation. The controls varying opening between fingers order allow move desired angular position following reference trajectory. We show experimentally...

We study the problem of robot interaction with mechanisms that afford one degree freedom motion, e.g., doors and drawers. propose a methodology for simultaneous compliant estimation constraints imposed by joint. Our method requires no prior knowledge mechanisms' kinematics, including type joint, prismatic or revolute. The consists velocity controller relies on force/torque measurements motion direction, distance, orientation rotational axis. It is suitable velocity-controlled manipulators...

Behavior Trees (BTs) were invented as a tool to enable modular AI in computer games, but have received an increasing amount of attention the robotics community last decade. With rising demands on agent complexity, game programmers found that Finite State Machines (FSM) they used scaled poorly and difficult extend, adapt reuse. In BTs, state transition logic is not dispersed across individual states, organized hierarchical tree structure, with states leaves. This has significant effect...

Modern industrial applications require robots to operate in unpredictable environments, and programs be created with a minimal effort, accommodate frequent changes the task. Here, we show that genetic programming can effectively used learn structure of behavior tree (BT) solve robotic task an environment. We propose use simple simulator for learning, demonstrate learned BTs same realistic simulator, converging without need specific heuristics, making our method appealing real applications.

Exoskeletons are increasingly used in rehabilitation and daily life patients with motor disorders after neurological injuries. In this paper, a realistic human knee exoskeleton model based on physical system was generated, human-machine created musculoskeletal modeling software, interactions different assistive strategies were simulated. The developed makes it possible to compute torques, muscle impulse, contact forces, interactive forces involved simulated movements. Assistive modeled as...

Human-robot collaboration (HRC) relies on accurate and timely recognition of human intentions to ensure seamless interactions. Among common HRC tasks, human-to-robot object handovers have been studied extensively for planning the robot's actions during reception, assuming intention handover. However, distinguishing handover from other has received limited attention. Most research focused visually detecting motion trajectories, which often results in delays or false detections when...

This work presents REFLEX: Robotic Explanations to FaiLures and Human EXpressions, a comprehensive multimodal dataset capturing human reactions robot failures subsequent explanations in collaborative settings. It aims facilitate research into human-robot interaction dynamics, addressing the need study both initial explanations, as well evolution of these long-term interactions. By providing rich, annotated data on responses different types failures, explanation levels, varying strategies,...