Robot manipulators should be able to quickly detect collisions to limit damage due to physical contact. Traditional model-based detection methods in robotics are mainly concentrated on the difference between the estimated and actual applied torque. In this paper, a model independent collision detection method is presented, based on the vibration features generated by collisions. Firstly, the natural frequencies and vibration modal features of the manipulator under collisions are extracted with illustrative examples. Then, a peak frequency based method is developed for the estimation of the vibration modal along the manipulator structure. The vibration modal features are utilized for the construction and training of the artificial neural network for the collision detection task. Furthermore, the proposed networks also generate the location and direction information about contact. The experimental results show the validity of the collision detection and identification scheme, and that it can achieve considerable accuracy.

This paper focuses on a new type of configuration design of a compliant parallel mechanism (CPM) planar continuum structure and its characteristic analysis of vibration-inherent frequency for planar motion, which can suppress the impact of random vibration in ultra-precision positioning and manufacturing equipment and improve the inherent frequency response of the mechanism. Firstly, a vector-mapping isomorphism between the fully CPM and conventional isomorphic parallel mechanism was constructed with a kinematic differential Jacobian matrix. Then, the mathematical model of topology optimization was put forward considering the compromise programming on the static stiffness and mean vibration-inherent frequency of the mechanism as the design variable and the minimization of compliance as the objective function. A constraint of volume fraction was considered and multi-objective micro displacement mechanism topology optimization based on a prismatic-revolute-revolute (3-PRR) planar nano-positioning continuum structure was performed using the solid isotropic material with penalization (SIMP) technique, which combines the criteria of the optimization algorithm and the vector isomorphic mapping method. Multi-objective topology optimization of the continuum structure micro displacement mechanism was investigated and presented by optimizations with different initial rejection rates. The simulation results show that the stiffness and vibration suppression performance of the continuum structure were improved, whereas the positioning of differential kinematics characteristics of the 3-PRR micro displacement planar fully CPM and isomorphic prototype mechanism retain the same. The modal analysis also provides a rational configuration for the micro displacement mechanism dimensional design and its optimal modal parameters. The crossover oscillation in frequency response of the continuum structure was reduced and quickly converged in the optimization iterations. The performance of the optimized mechanism was verified by the experiments on a planar fully compliant micro displacement continuum structure based on Lead Zirconate Titanate (PZT) actuator.

For the robotic grasping of randomly stacked objects in a cluttered environment, the active multiple viewpoints method can improve grasping performance by improving the environment perception ability. However, in many scenes, it is redundant to always use multiple viewpoints for grasping detection, which will reduce the robot’s grasping efficiency. To improve the robot’s grasping performance, we present a Viewpoint Adjusting and Grasping Synergy (VAGS) strategy based on deep reinforcement learning which coordinates the viewpoint adjusting and grasping directly. For the training efficiency of VAGS, we propose a Dynamic Action Exploration Space (DAES) method based on $\epsilon$-greedy to reduce the training time. To address the sparse reward problem in reinforcement learning, a reward function is created to evaluate the impact of adjusting the camera pose on the grasping performance. According to experimental findings in simulation and the real world, the VAGS method can improve grasping success and scene clearing rate. Compared with only direct grasping, our proposed strategy increases the grasping success rate and the scene clearing rate by 10.49% and 11%.

To solve the problem that the time-consuming optimization process of Genetic Algorithm (GA) can erode the expected time-saving brought by the algorithm, time-optimal trajectory planning based on cubic spline was used, after the modification to classical fitness sharing function of NGA, a dual-threaded method utilizing elite strategy characteristic was designed which was based on Niche Genetic Algorithm (NGA) with the fitness sharing technique. The simulation results show that the proposed method can mitigate the contradiction of the long term the optimization algorithm takes but a short running time the trajectory gets, demonstrating the effectiveness of the proposed method. Besides, the improved fitness sharing technique has reduced the subjective process of determining relevant parameters and the optimized trajectory results met performance constraints of the robot joints.

This paper proposes a safety-constrained interaction control scheme for robotic manipulators by integrating model predictive control (MPC) and active disturbance rejection control (ADRC). The proposed method is specifically designed for manipulators with complex nonlinear dynamics. To ensure that the control system satisfies safety constraints during human–robot interaction, MPC is incorporated into the impedance control framework to construct a model predictive impedance controller (MPIC). By exploiting the prediction and constraint-handling capabilities of MPC, the controller provides guaranteed safety throughout the interaction process. Meanwhile, ADRC is employed to track the target joint control signals generated by the MPIC, where an extended state observer is utilized to compensate for dynamic modeling errors and nonlinear disturbances within the system, thereby achieving accurate trajectory tracking. The proposed method is validated through both simulation and real-world experiments, achieving high-performance interaction control with safety constraints at a 2 ms control cycle. The controller exhibits active compliant interaction behavior when the interaction stays within the constraint boundaries, while maintaining strict adherence to the safety constraints when the interaction tends to violate them.

Smart factories have real-time demands for the statistics of productivity to meet the needs of quick reaction capabilities. To solve this problem, a counting method based on our decomposition strategy of actions was proposed for complex actions. Our method needs to decompose complex actions into several essential actions and define a label string for each complex action according to the sequence of the essential actions. While counting, we firstly employ an online action recognition algorithm to transform video frames into label numbers, which will be stored in a result queue. Then, the label strings are searched for their results in queue. If the search succeeds, a complex action will be considered to have occurred. Meanwhile, the corresponding counter should be updated to accomplish counting. The comparison test results in a video dataset of workers’ repetitive movements in package printing production lines and illustrate that our method has a lower counting errors, MAE (mean absolute error) less than 5% as well as an OBOA (off-by-one accuracy) more than 90%. Moreover, to enhance the adaptability of the action recognition model to deal with the change of action duration, we propose an adaptive parameter module based on the Kalman filter, which improves counting performances to a certain extent. The conclusions are that our method can achieve high counting performance, and the adaptive parameter module can further improve performances.