Yuichi Kobayashi

Conventional robot motion teaching methods are time-consuming and place a significant burden on operators, especially when trajectory modification is required after contact with the environment. This paper proposes a six-dimensional automatic path correction method that considers both the position and orientation of the end-effector. The operator specifies only the start, goal, and via points of the motion path. Contact wrench information obtained during execution is incorporated into a cost function, and the control points of a spline-based trajectory are iteratively optimized. The method does not rely on an explicit environment model and autonomously updates the trajectory based on contact information. Simulation experiments were conducted on three industrial tasks: part box unloading, yarn cone removal, and part extraction. In all tasks, the proposed method successfully generated collision-free trajectories within a small number of correction iterations and with practical computation time. These results indicate that the proposed approach enables stable and efficient trajectory correction while reducing the teaching burden in industrial support tasks.

Developing a learning model that adapts to changes in the body is critical for improving the flexibility of machine intelligence. During recovery from a controller malfunction, humans use the information obtained from previous experiences. One possible explanation for the recovery process is that information from the remaining controller was transformed and used. Modeling this mechanism will aid in the development of an adaptive motor-learning model capable of quickly recovering from controller malfunctions. We proposed a learning model for explaining the reused information of the remaining controllers in a pair of controllers. Simulations of a pair of upper limbs validated that the learning model could find a simple transformation, such as a reflection between the left and right arms, using optimization.

In cases where vision is not sufficiently reliable for robots to recognize an object, tactile sensing can be a promising alternative for estimating the object’s pose. In this paper, we consider the task of a robot estimating the pose of a container aperture in order to select an object. In such a task, if the robot can determine whether its hand with equipped contact sensor is inside or outside the container, estimation of the object’s pose can be improved by reflecting the discrimination to the robotic hand’s exploration strategy. We propose an exploration strategy and an estimation method using discrete state recognition on the basis of a particle filter. The proposed method achieves superior estimation in terms of the number of contact actions, operation time, and stability of estimation efficiency. The pose is estimated with sufficient accuracy that the hand can be inserted into the container.

To achieve robot navigation in crowded environments having high densities of moving people, it is insufficient to simply consider humans as moving obstacles and avoid collisions with them. That is, the impact of an approaching robot on human movements must be considered as well. Moreover, various navigation methods have been tested in their own environments in the literature, which made them difficult to compare with one another. Thus, we propose an autonomous robot navigation method in densely crowded environments for data-based predictions of robot-human interactions, together with a reproducible experimental test under controlled conditions. Based on localized positional relationships with humans, this method extracts multiple alternative paths, which can implement either following or avoidance, and selects an optimal path based on time efficiency. Each path is selected using neural networks, and the various paths are evaluated by predicting the position after a given amount of time has elapsed. These positions are then used to calculate the time required to reach a certain target position to ensure that the optimal path can be determined. We trained the predictor using simulated data and conducted experiments using an actual mobile robot in an environment where humans were walking around. Using our proposed method, collisions were avoided more effectively than when conventional navigation methods were used, and navigation was achieved with good time efficiency, resulting in an overall reduction in interference with humans. Thus, the proposed method enables an effective navigation in a densely crowded environment, while collecting human-interaction experience for further improvement of its performance in the future.

The exploration of remote, unknown, rough environments by autonomous robots strongly depends on the ability of the on-board system to build an accurate predictor of terrain traversability. Terrain traversability prediction can be made more cost efficient by using texture information of 2D images obtained by a monocular camera. In cases where the robot is required to operate on a variety of terrains, it is important to consider that terrains sometimes contain spiky objects that appear as non-uniform in the texture of terrain images. This paper presents an approach to estimate the terrain traversability cost based on terrain non-uniformity detection (TNUD). Terrain images undergo a multiscale analysis to determine whether a terrain is uniform or non-uniform. Terrains are represented using a texture and a motion feature computed from terrain images and acceleration signal, respectively. Both features are then combined to learn independent Gaussian Process (GP) predictors, and consequently, predict vibrations using only image texture features. The proposed approach outperforms conventional methods relying only on image features without utilizing TNUD.

In order to make an autonomous robot system more adaptive to human-centered environments, it is effective to let the robot collect sensor values by itself and build controller to reach a desired configuration autonomously. Multiple sensors are often available to estimate the state of the robot, but they contain two problems: (1) sensing ranges of each sensor might not overlap with each other and (2) sensor variable can contain redundancy against the original state space. Regarding the first problem, a local coordinate definition based on a sensor value and its extension to unobservable region is presented. This technique helps the robot to estimate the sensor variable outside of its observation range and to integrate regions of two sensors that do not overlap. For a solution to the second problem, a grid-based estimation of lower-dimensional subspace is presented. This estimation of manifold allows the robot to have a compact representation, and thus the proposed motion generation method can be applied to the redundant sensor system. In the case of image feature spaces with a high-dimensional sensor signal, a manifold estimation-based mapping, known as locally linear embedding (LLE), was applied to an estimation of distance between robot body and an obstacle.

This paper presents an action decision framework for an autonomous mobile robot or an unmanned ground vehicle (UGV) to navigate an unknown environment. It is difficult for a UGV without global map information to decide which path to travel when it comes to a fork. However, locally observed terrain features can enable the UGV if it can utilize its past experience. The proposed path selection method utilizes correlations between features of the local terrain obtained by its laser range finder and the values of paths obtained through offline simulation using global path planning. During navigation, the UGV estimates the values of each path at a fork based on the correlation between the terrain feature and the value. It was confirmed that the proposed method allows the selection of paths that are more effective compared with a simple path selection strategy with which the UGV selects the closer path to the goal. The proposed method was evaluated in both a simulated environment and a real outdoor environment.

Because of the frequent occurrence of large-scale disasters, such as earthquakes, tsunamis, volcanic eruptions, and river floods, there is an increased demand for emergency response, restoration, and disaster prevention using robotic technology. One such technology involves assessment of the damage status using flying robots, which have undergone rapid development in recent years. In this study, using images of the disaster site obtained from a flying robot and a terrain database consisting of predisaster 3-D data, we aim to detect efficiently the collapse of electric utility poles, which are man-made objects, and the water level difference before and after river flooding, which is part of the natural landscape. By detecting these disaster-related events, we show the validity of the proposed method to assess the damage situation using the terrain database.

It is expected that telepresence techniques are introduced into a car navigation system. For example, the navigation system presents a road scene taken from different points of view to drivers. The realization of the system require cameras on a road demanded by drivers. However, available cameras of roads are generally insufficient. We aim to construct an another view point image generation with shifted perspective using two in-vehicle camera images based on ad-hoc networks. Using the models of a few virtual planes produced from measurement points, we have shown that the transformation is performed in real time. Experimental results demonstrated the effectiveness of the proposed method.

Autonomous robots that work in the same environment as humans must operate safely and adapt to handle various tools and deal with partial malfunctions. We propose an approach for estimating the robot structure and apply this approach for building a controller of dynamic motions. The robot structure is estimated by evaluating the mutual information (MI) among the sensor variables. Variables with high values of MI are edge-connected and the controller is automatically constructed based on the estimated structure. The proposed approach can accommodate changes in the robot parameters and dynamic motions. We verify the proposed method by using a simulator of a musculoskeletal arm driven that is driven by artificial muscle for mechanical safety.