In the field of lower limb exoskeletons, it is essential to accurately estimate the gait phase of humans. Many methods have been proposed to estimate the gait phase, but only a few studies have considered the multi-locomotion mode. This paper proposes a novel inertial measurement unit(IMU)-based method to estimate the gait phase of a pilot in continuous multi-locomotion mode. The method includes gait pattern recognition based on long short-term memory (LSTM), continuous phase estimation based on a dual adaptive frequency oscillator(DAFO), threshold-based toe-off event detection and a rule-based gait phase synchronization module. First, we used the LSTM-based network to identify four gait patterns including standing, level ground walking, upstairs and downstairs. Next, the DAFO was used to obtain the continuous gait phase of the pilot. Then, we detected the gait events in different gait modes. Finally, the continuous gait phase was synchronized according to the gait events. The experimental result shows that the gait pattern classification accuracy using 5 IMUs is 98.58% and the $\mathcal {F}_{\!{1}}$ score reaches 0.9875. The proposed DAFO model can maintain good stability when multiple gait modes are frequently switched, significantly improving the problem of slow convergence and the poor robustness of single adaptive frequency oscillator(SAFO) models. Toe-off gait events of 492 steps are all detected and the average error at the detected gait events in different gait modes is 15.34± 40.58 ms.
31 July 2023 in IEEE Robotics and Automation Letters
Accurate recognition of human pose and prediction of human motion intention are essential for exoskeleton robots to provide effective assistance. In recent years, inertial measurement units (IMUs) have been widely used to estimate human pose due to their stability and economy. However, physical sensors are inherently subject to time delay relative to human movement, which makes it challenging to meet the demand for real-time control of power-assisted exoskeletons. To address this issue, we propose a data-driven approach called motion forecasting network (MoFCNet) that leverages historical IMU readings to forecast the 3D pose of human in future frames. Our approach consists of two stages, state variables forecasting and pose regression. In stage I, we forecast the IMU data, joint positions and velocities. In stage II, we regress the future human poses from the forecasting results of stage I. Besides, we design a basic stack that can be applied to both stages. The mean joint rotation error on DIP-IMU and TotalCapture datasets are $\bm {10.34}$ and $\bm {16.11}$ degrees, respectively. When focusing solely on the first frame of the prediction results, these errors decrease to 9.13 and 14.50 degrees. Experimental results indicate that our method not only outperforms the forecasting baseline methods but also achieves comparable performance to the sparse IMU-based human pose estimation algorithm.
2 August 2023 in IEEE Robotics and Automation Letters
We introduce a new task, called Class Incremental Robotic Pick-and-Place (CIRPAP), which calls for the capacity to learn to pick and place new categories of objects while retaining the skill of dealing with the previously learned ones. CIRPAP faces three challenges: catastrophic forgetting, few-shot learning, and robust picking in cluttered environments. To address the challenges of catastrophic forgetting and few-shot learning, we propose a novel CIRPAP framework that is built on Incremental Few-Shot Object Detection (iFSD). Specifically, with fixed pre-trained Transformer-like object detection models, we only fine-tune the additional adapter modules, which is called adapter-tuning. To address the challenge of robust picking in cluttered environments, we also utilize multiview fusion to integrate object detection and grasp prediction results. As for iFSD evaluation, experiments show that our adapter-tuning-based approach outperforms state-of-the-art methods on COCO and our dataset. As for full CIRPAP system evaluation, experimental results on a real robotic platform demonstrate the effectiveness of our proposed framework.
In recent years, parallel robots have become a hot research topic in trauma fracture treatment because of their high precision, high load capacity, and compact structure. However, parallel robots have disadvantages like small workspaces and complex singularity. In this article, a novel redundant parallel mechanism (RPM) for long bone fracture reduction is proposed based on Stewart parallel mechanism (SPM). Six kinematically redundant DOFs (degrees-of-freedom) are added to the RPM. First, the kinematics of the RPM is established, and its workspace is calculated. The analysis results indicate that the position workspace of the RPM is about 19 times larger than that of the SPM. The RPM has a similar range of torsion angles as the SPM, but a more extensive range of tilt angles than the SPM. Second, the singularities of the two parallel mechanisms are compared based on the dimensionally homogeneous Jacobian matrix. The results show that the dexterity of the RPM is much better than the SPM. Third, a multiparameter multi-objective optimization method is proposed to optimize the geometry parameters of the RPM. The statics of the RPM is analyzed by finite element analysis. To further expand the performance of the RPM, the unfixed RPM (URPM) is proposed. The analysis results show that the URPM is superior to the RPM in terms of workspace and dexterity. Finally, experiments are conducted to verify the effectiveness of the proposed methods in this article.
1 September 2018 in International Journal of Advanced Robotic Systems
This article presents a strategy for efficient sorting path planning and trajectory optimization with multiple constraints. The proposed strategy is applicable to typical high-density sorting applications. It plans the sorting path while optimizing each trajectory. The total stroke traveled by the end effector to complete the sorting task is shortened, and the efficiency of the production line is correspondingly facilitated. Thus, this article designs bow-shaped trajectories by analyzing the work process of the production line to ensure the smooth transition of the end effector and bow-shaped trajectories are based on the starting point and the end point of each sorting action. The modified trapezoidal acceleration and deceleration algorithm is used to optimize each sorting trajectory, which ensures that each sorting action is executed quickly and stably. Simultaneously, the greedy strategy is applied to plan the sorting task, which further shortens the total stroke of the end effector and improves the efficiency of the production line. A specific cost function is also designed to improve the planning strategy to prevent the occurrence of missing materials and enhance the adaptability of the sorting system. In consideration of the efficiency of the sorting system is significantly improved by this approach, the effectiveness of the proposed strategy in this article is thus verified compared with existing ones by experiments. In addition, the impact of the conveyor speed on the sorting system is also discussed.