Recurrent Neural Networks-Based Collision-Free Motion Planning for Dual Manipulators Under Multiple Constraints

Abstract

Dual robotic manipulators are robotic systems that are developed to imitate human arms, which shows great potential in performing complex tasks. Collision-free motion planning in real time is still a challenging problem for controlling a dual robotic manipulator because of the overlap workspace. In this paper, a novel planning strategy under physical constraints of dual manipulators using dynamic neural networks is proposed, which can satisfy the collision avoidance and trajectory tracking. Particularly, the problem of collision avoidance is first formulated into a set of inequality formulas, whereas the robotic trajectory is then transformed into an equality constraint by introducing negative feedback in outer loop. The planning problem subsequently becomes a Quadratic Programming (QP) problem by considering the redundancy, the boundaries of joint angles and velocities of the system. The QP is solved using a convergent provable recurrent neural network that without calculating the pseudo-inversion of the Jacobian. Consequently, numerical experiments on 8-DoF modular robot and 14-DoF Baxter robot are conducted to show the superiority of the proposed strategy. INDEX TERMS Motion planning, dual robotic manipulators, dynamic neural networks, zeroing neural networks, redundant resolution. JINGLUN LIANG was born in Meizhou, China, in 1985. He received the B.E. degree in mechatronic engineering and the Ph.D. degree in mechanical manufacturing and automation from the

Augmented Reality-Based Interface for Bimanual Robot Teleoperation

“…The vast majority of contributions in bimanual robotics present fully automated tasks based, for instance, on artificial intelligence techniques [21,23], motion planing techniques [14,15,26,32,33], or other low-level control approaches [19,24,25,27,28].…”

Collision avoidance trajectory planning for a dual-robot system: using a modified APF method

“…In aspect of the intelligent optimization algorithm-based method, Liang et al [11] proposed a dual-robot collision-free motion planning strategy based on a dynamic neural network with physical constraints, which can successfully avoid collision. By taking this approach, the planning errors of 8-DOF modular robot and 14-DOF Baxter robot are reduced to 0 within 0.5 s and 0.3 s, respectively.…”

Design of detection‐jamming shared waveform based on virtual force field algorithm

“…Simultaneously, in order to solve the time-domain conflict among detection-jamming shared signals, its echo signals, and reconnaissance signals, the multidimensional parameter characteristics of the false target of the detection-jamming shared signals are optimized by introducing and taking full use of the advantages of the virtual force field algorithm (VFFA), 21,22 such as simple operation, fast convergence speed, and real-time obstacle avoidance. [23][24][25][26][27] Then, the separation in the time sequence among reconnaissance, jamming, and detection is achieved effectively by designing the detection-jamming shared signal based on VFFA, and a good deceptive effect is shown on the noncollaborative target's radar screen receiver. After the related signal processing of the integrated detection-jamming system, the echo signal can achieve excellent nonambiguous distance, as well as velocity measurements, and has a good effect of pulse accumulation, showing good detection performance.…”

“…Further, its waveform characteristics are similar to the noncollaborative pulse. Simultaneously, in order to solve the time‐domain conflict among detection‐jamming shared signals, its echo signals, and reconnaissance signals, the multidimensional parameter characteristics of the false target of the detection‐jamming shared signals are optimized by introducing and taking full use of the advantages of the virtual force field algorithm (VFFA), 21,22 such as simple operation, fast convergence speed, and real‐time obstacle avoidance 23–27 . Then, the separation in the time sequence among reconnaissance, jamming, and detection is achieved effectively by designing the detection‐jamming shared signal based on VFFA, and a good deceptive effect is shown on the noncollaborative target's radar screen receiver.…”