Search

Recently, research has been conducted on mixed reality (MR), which provides immersive visualization and interaction experiences, and on mapping human motions directly onto a robot in a mixed reality (MR) space to achieve a high level of immersion. However, even though the robot is mapped onto the MR space, their surrounding environment is often not mapped sufficiently; this makes it difficult to comfortably perform tasks that require precise manipulation of the objects that are difficult to see from the human perspective. Therefore, we propose a system that allows users to operate a robot in real space by mapping the task environment around the robot on the MR space and performing operations within the MR space.

With the increasing demand for human–robot collaboration (HRC), intuitive interfaces are essential to connect humans and robots. A promising approach is the use of mixed reality (MR) to enhance spatial understanding through virtual and augmented reality. In this paper, we propose a novel HRC system that extracts human handling procedures and generates concrete motion plans for the robot. The user, wearing an MR device, interacts with virtual objects in the MR space using natural hand motions. These motions and resulting state transitions are abstracted into a symbolic semi-order motion planner represented by the reachability graph (RG). Using the RG, an autonomous behavior tree is generated, considering the robot’s task environment, and the concrete motion plan is executed by the robot. This system allows the robot to take a more flexible approach to user instructions than conventional MR-HRC systems. Moreover, this system translates human orders into plans that are independent of a specific robot, demonstrating considerable development potential.

The growing need for high levels of autonomy in Autonomous Robotic Surgery Systems (ARSS) calls for innovative approaches to reduce surgeons’ cognitive load, optimize hospital workflows, and ensure efficient task-level reasoning and adaptation during execution. This paper presents a novel hybrid framework that synergistically combines Task-Motion Planning and Dynamic Behavior Trees for ARSS in Minimally Invasive Surgery. Our approach is designed to address the challenges of coordinating multiple surgical tools within a small workspace, thereby making complex surgical tasks like multi-throw suturing feasible and efficient. Through an extensive evaluation in simulation across diverse initial conditions and noise scenarios, the proposed method demonstrates improved success rates, reduced execution times, and fewer regrasps compared to standalone approaches. Furthermore, it showcases robustness under increased noise conditions. By applying our framework to a complex multi-throw suturing task, we illustrate its capability to seamlessly handle comprehensive suturing tasks, including needle picking, insertion, extraction, and the handover of the needle between Patient Side Manipulators. The results suggest that our hybrid approach not only enhances ARSS autonomy but also adapts effectively to unexpected environmental changes, laying the groundwork for its potential applicability in real-world surgical robotics.

This study processes an autonomous indoor drone photographer that searches for and selects a heuristic optimal viewpoint to obtain a well-composed photograph of a group of subjects. The subjects on the drone’s camera screen are represented by a Gaussian mixture model. When there are four or more subjects, they are represented by a Gaussian mixture model with clustering by variational Bayes. The Kullback–Leibler divergence is evaluated between the Gaussian mixture model and a user-defined reference composition, and it is defined as the composition evaluation value. The reference composition is pre-set by the user based on the basic composition rules, such as the three-section method. The drone searches for a viewpoint in a 3D space to optimize the composition evaluation value using particle swarm optimization (PSO). A Gaussian process is used to facilitate the PSO search. This enables the drone to significantly reduce the search time and successfully capture a photograph with a well-balanced composition.

Abstract This paper addresses a reference gait speed generator for limit-cycle-based bipedal gait of humanoid robots in order to achieve secure and efficient acceleration and deceleration without falling down at respective gait speeds. The reference gait speed generator is hard to be developed by analytical approaches due to the necessity of barren work for identifying the respective basins of attraction. We challenge this issue by designing virtual dynamics among a robot, a virtual leader point, and a goal, and adapting it according to a falling risk of the robot. The virtual dynamics, which has settling and acceleration times as design parameters, gives the reference speeds derived from states of the robot and the leader point to a gait speeds controller. In the dynamics, the robot’s mass is optimized virtually to maximize efficiency while ensuring stability stochastically by using a selection algorithm for locomotion. Even when there were obstacles or an up slope in traveling courses of simulations, the robot achieved the autonomous traveling from the start to the goal securely. Specific resistance was also kept small in comparison with local-stability-based walking. The proposed method makes the limit-cycle-based bipedal gait more practical and contributes toward replacing the major method that ensures stability of every step.

Abstract A visual assistance system has become attractive as a technique to improve the efficiency and stability of remote control. While an operator controls a working robot, another autonomous monitoring robot evaluates a suitable viewpoint to observe the work site, and dynamically moves to the optimal viewpoint for monitoring. Choosing the observation region (ROI: region of interest) is equivalent to deciding the action of the following autonomous monitoring system. Therefore, we focus on ROI detection in our visual support system. We propose an ROI selection method to identify the most suitable observation point and interobject relations. The monitoring robot detects a gestalt of the scene in order to identify the relations between objects. Such an adaptive ROI in real time improves the efficiency of the remote control. The experimental results indicate the effectiveness of the proposed system in terms of execution time and number of errors.

This paper aims to determine an optimal structure for a quadruped robot, which will allow the robot’s joint torque sum to be minimised. An animal’s characteristic limb length ratio is a vital part of its overall morphology and the one that enables it to travel easily through its environment. For the same reason, a robot’s structure needs to be suitably designed for locomotion in its working environment. Joint torques are necessary to maintain the posture of the robot and to accelerate joint angles during walking motion, hence, minimisation of joint torques reduces energy consumption. We performed a numerical simulation in which we analysed the joint torques for various limb lengths and slope angles in order to determine the optimal structure of a robot walking on a slope. Our investigation determines that the optimal Ratio of Rear Leg Length (RRL) can be derived by the use of a simulation designed to determine the physical structure of quadruped robot. Our analysis suggests that joint torque will increase as the slope angle becomes steeper if the rear legs of the robot are shorter than its forelegs, and that joint torque will decrease as the slope angle declines if the robot’s forelegs are shorter than its rear legs. Finally, experimental results validated our simulation analysis.

Bipedal walking ability, which is mainly composed of mechanical stability and moving efficiency, is one of the most important issues in the field of humanoid robots. An effective use of the arms is expected to improve the walking ability under the constraints from limited body. Multiple usages of the arms are required to adapt the robot to various walking conditions because each usage possesses different utility, respectively. We propose two strategies for arm-swing to enhance stability and efficiency. Both the strategies use the selection algorithm for locomotion (SAL) that determines the optimal locomotion for the various walking conditions. One strategy is a use/compensation of moment of swing-leg by hip-rotation/arm-swing. This strategy, called Ro-SAL, selects a rotation rate of hip depending upon a falling risk. Another strategy is a support of the center of gravity (COG) tracking. This strategy, called Su-SAL, employs a predictive control, and a support weight that determines weights of input and output in the predictive control is also selected depending upon the falling risk. From simulation results, we validate that Ro-SAL is effective in a stable state and a state with internal model error, while Su-SAL is effective in states with external force and environmental complexity.

Gesture recognition is essential for human and robot collaboration. Within an industrial hybrid assembly cell, the performance of such a system significantly affects the safety of human workers. This work presents an approach to recognizing hand gestures accurately during an assembly task while in collaboration with a robot co-worker. We have designed and developed a sensor system for measuring natural human-robot interactions. The position and rotation information of a human worker's hands and fingertips are tracked in 3D space while completing a task. A modified chain-code method is proposed to describe the motion trajectory of the measured hands and fingertips. The Hidden Markov Model (HMM) method is adopted to recognize patterns via data streams and identify workers' gesture patterns and assembly intentions. The effectiveness of the proposed system is verified by experimental results. The outcome demonstrates that the proposed system is able to automatically segment the data streams and recognize the gesture patterns thus represented with a reasonable accuracy ratio.