Detecting driver fatigue is critical to ensuring road safety. Existing fatigue detection methods typically rely on traditional hand-picked features as inputs. However, these hand-picked features can hardly respond accurately to the driver’s fatigue state due to a certain degree of subjectivity and the extraction of these features requires a long time window, which limits the accuracy and real-time performance of the detection. This paper proposes a novel fatigue detection method based on multi-source information fusion, which relies entirely on neural networks for automatic feature extraction. Through simulated driving experiments, we recorded physiological signals and facial videos from 21 participants for model training and testing. The results show that our model outperforms existing methods in terms of accuracy and real-time performance, achieving a detection accuracy of 93.15% within a 3-second time window (specificity = 94.04%, sensitivity = 91.71%). The visualization results of the model reveal potential relationships between facial regions for the first time, validating the rationality and effectiveness of our method. The practical issues of fatigue detection methods and future research directions are also explored.

In this paper, we propose an approach to design a proper fixture layout for a 3D curved workpiece. The problem is tackled in a point set domain by discretizing the exterior surface of the workpiece into a dense collection of candidate fixturing points and then searching for a small proper set of fixturing points that satisfies the total restraint of the workpiece and reduces the workpiece positioning error. The algorithm first randomly selects an initial set of seven fixturing points and then iteratively improves it by exchanging with the other candidate points through a best-first strategy and a randomized motion. Finally, the algorithm has been implemented and its efficiency has been ascertained by two examples.

The calibration accuracy of the intrinsic and extrinsic parameters of the vision system greatly affects the performance of visual servoing. We address the problem of controlling a planar manipulator using a fixed single camera without calibrating its intrinsic parameters and the transformation matrix between the robot base frame and the camera frame, and without measuring manipulator's depth. Based on an important observation that the unknown parameters can be separated from the unknown composite image Jacobian matrix, we propose an adaptive algorithm to estimate the unknown and mixed parameters on-line. It is proved with a full consideration of dynamics of the system by Lyapunov approach that the feature points of planar manipulator approach asymptotically to the desired ones on image plane and the estimated parameters are bounded under the control of the proposed visual servo controller. The performance has been confirmed by simulations and experiments.

The paper proposes a new haptic system for the pulse palpation via the Internet for remote diagnosis of patients without time and space restrictions. First, we design a pulse-detect sensor to obtains the pulse signals from the patient's wrist. Next, a server-client process is developed to transfer the pulse signals to the remote computer. Then, a haptic device is used to provide an interface for the doctor to feel the pulses of the patient at a remote distance. The haptic device regenerates the pulse signals by a motor-controlled cam system. To couple the doctor with the remote environment and give him a sense of telepresence, a trajectory follower is implemented: the cam follower moves up and down in accordance with the desired pulse trajectory. In addition, by considering the system as a one-dimensional manipulator, a PID force controller is realized to trace the desired force obtained from pressure sensors on the patient's side. As a result, the force felt by the doctor from the haptic device is closed to that he/she would feel from the patient's wrist directly.

This paper presents a simple and efficient algorithm to calculate the fingertip positions which ensure a form-closure grasp. The algorithm first arbitrarily chooses a grasp on the given faces of the object. If the selected grasp is not form-closure, the origin O of the wrench space lies outside of the convex hull of the primitive contact wrenches. In this case, the algorithm properly moves the fingertip positions at a fixed step on the faces so that the primitive contact wrenches move towards the origin O until the origin is eventually contained by the convex hull after iterations of the operation. The form-closure property is checked by our ray-shooting based qualitative test algorithm. The motion of the fingertip positions at every step is determined by a quadratic programming problem. Finally we have implemented the proposed algorithm and verified its efficiency with three numerical examples.