As there come to be more applications of intelligent robots, their task object is becoming more varied. However, it is still a challenge for a robot to handle unfamiliar objects. We review the recent work on the feature sensing and robotic grasping of objects with uncertain information. In particular, we focus on how the robot perceives the features of an object, so as to reduce the uncertainty of objects, and how the robot completes object grasping through the learning-based approach when the traditional approach fails. The uncertain information is classified into geometric information and physical information. Based on the type of uncertain information, the object is further classified into three categories, which are geometric-uncertain objects, physical-uncertain objects, and unknown objects. Furthermore, the approaches to the feature sensing and robotic grasping of these objects are presented based on the varied characteristics of each type of object. Finally, we summarize the reviewed approaches for uncertain objects and provide some interesting issues to be more investigated in the future. It is found that the object’s features, such as material and compactness, are difficult to be sensed, and the object grasping approach based on learning networks plays a more important role when the unknown degree of the task object increases.

The tiger beetle can maintain its stability by controlling the interaction force between its legs and an unstructured terrain while it runs. The biomimetic hexapod robot mimics a tiger beetle, and a comprehensive force sensing system combined with certain algorithms can provide force information that can help the robot understand the unstructured terrain that it interacts with. This study introduces a complicated leg force sensing system for a hexapod robot that is the same for all six legs. First, the layout and configuration of sensing system are designed according to the structure and sizes of legs. Second, the joint toque sensors, 3-DOF foot-end force sensor and force information processing module are designed, and the force sensor performance parameters are tested by simulations and experiments. Moreover, a force sensing system is implemented within the robot control architecture. Finally, the experimental evaluation of the leg force sensor system on the hexapod robot is discussed and the performance of the leg force sensor system is verified.

After each robot end tool replacement, tool center point (TCP) calibration must be performed to achieve precise control of the end tool. This process is also essential for robot-assisted puncture surgery. The purpose of this article is to solve the problems of poor accuracy stability and strong operational dependence in traditional TCP calibration methods and to propose a TCP calibration method that is more suitable for a physician. This paper designs a special binocular vision system and proposes a vision-based TCP calibration algorithm that simultaneously identifies tool center point position (TCPP) and tool center point frame (TCPF). An accuracy test experiment proves that the designed special binocular system has a positioning accuracy of ±0.05 mm. Experimental research shows that the magnitude of the robot configuration set is a key factor affecting the accuracy of TCPP. Accuracy of TCPF is not sensitive to the robot configuration set. Comparison experiments show that the proposed TCP calibration method reduces the time consumption by 82%, improves the accuracy of TCPP by 65% and improves the accuracy of TCPF by 52% compared to the traditional method. Therefore, the method proposed in this article has higher accuracy, better stability, less time consumption and less dependence on the operations than traditional methods, which has a positive effect on the clinical application of high-precision robot-assisted puncture surgery.

Inspired by visual-tactile cross-modal bidirectional mapping of the human brain, this paper introduces a novel approach to bidirectional mapping between visual and tactile data, an area not fully explored in the predominantly unidirectional existing studies. First, we adopt separate Variational AutoEncoder (VAE) models for visual and tactile data. Furthermore, we introduce a conditional flow model built on the VAE latent feature space, enabling cross-modal bidirectional mapping between visual and tactile data using one model. The experimental results show that our method achieves excellent performance in terms of the similarity between the generated data and the original data (Structural Similarity Index (SSIM) of visual data: 0.58, SSIM of tactile data: 0.80), the classification accuracy on generated data (visual data: 91.60%, tactile data: 88.05%), and the zero-shot classification accuracy between generated data and language (visual data: 44.49%, tactile data: 45.03%). To the best of our knowledge, the method proposed in this paper is the first one to utilize a single model to achieve bidirectional mapping between visual and tactile data. Our model and code will be made public after the acceptance of the paper.

In the human–robot collaboration system, the high-precision distortion correction of the camera as an important sensor is a crucial prerequisite for accomplishing the task. The traditional correction process is to calculate the lens distortion with the camera model parameters or separately from the camera model. However, in the optimization process calculate with the camera model parameters, the mutual compensation between the parameters may lead to numerical instability, and the existing distortion correction methods separated from the camera model are difficult to ensure the accuracy of the correction. To address this problem, this study proposes a model-independent lens distortion correction method based on the image center area from the perspective of the actual camera lens distortion principle. The proposed method is based on the idea that the structured image preserves its ratios through perspective transformation, and uses the local image information in the central area of the image to correct the overall image. The experiments are verified from two cases of low distortion and high distortion under simulation and actual experiments. The experimental results show that the accuracy and stability of this method are better than other methods in training and testing results.

In manufacturing, traditional task pre-programming methods limit the efficiency of human–robot skill transfer. This paper proposes a novel task-learning strategy, enabling robots to learn skills from human demonstrations flexibly and generalize skills under new task situations. Specifically, we establish a markerless vision capture system to acquire continuous human hand movements and develop a threshold-based heuristic segmentation algorithm to segment the complete movements into different movement primitives (MPs) which encode human hand movements with task-oriented models. For movement primitive learning, we adopt a Gaussian mixture model and Gaussian mixture regression (GMM-GMR) to extract the optimal trajectory encapsulating sufficient human features and utilize dynamical movement primitives (DMPs) to learn for trajectory generalization. In addition, we propose an improved visuo-spatial skill learning (VSL) algorithm to learn goal configurations concerning spatial relationships between task-relevant objects. Only one multioperation demonstration is required for learning, and robots can generalize goal configurations under new task situations following the task execution order from demonstration. A series of peg-in-hole experiments demonstrate that the proposed task-learning strategy can obtain exact pick-and-place points and generate smooth human-like trajectories, verifying the effectiveness of the proposed strategy.