Multi-Objective Optimal Trajectory Planning for Robotic Arms Using Deep Reinforcement Learning.

This study investigated the trajectory-planning problem of a six-axis robotic arm based on deep reinforcement learning. Taking into account several characteristics of robot motion, a multi-objective optimization approach is proposed, which was based on the motivations of deep reinforcement learning and optimal planning. The optimal trajectory was considered with respect to multiple objectives, aiming to minimize factors such as accuracy, energy consumption, and smoothness. The multiple objectives were integrated into the reinforcement learning environment to achieve the desired trajectory. Based on forward and inverse kinematics, the joint angles and Cartesian coordinates were used as the input parameters, while the joint angle estimation served as the output. To enable the environment to rapidly find more-efficient solutions, the decaying episode mechanism was employed throughout the training process. The distribution of the trajectory points was improved in terms of uniformity and smoothness, which greatly contributed to the optimization of the robotic arm's trajectory. The proposed method demonstrated its effectiveness in comparison with the RRT algorithm, as evidenced by the simulations and physical experiments.

A novel transgenic zebrafish line allows for in vivo quantification of autophagic activity in neurons.

The pathophysiology of most neurodegenerative diseases includes aberrant accumulation of protein aggregates. Recent evidence highlights the role of protein degradation pathways in neurodegeneration. Concurrently, genetic tools have been generated to enable zebrafish, Danio rerio, to be used as an animal model to study neurodegenerative processes. In addition to optical clarity and fast ex utero development, the zebrafish brain is relatively small and has conserved structures with its mammalian counterparts. To take advantage of this model organism and to aid further studies on autophagy and neurodegeneration, we created a stable transgenic zebrafish line that expresses eGFP-Map1lc3b specifically in post-mitotic neurons under the elavl3 promoter. This line is useful for indirectly monitoring autophagic activity in neurons in vivo and screening for macroautophagy/autophagy-modulating compounds. We determined the applicability of this transgenic line by modulating and quantifying the number of autophagosomes via treatment with a known autophagy inducer (rapamycin) and inhibitors (3-methyladenine, protease inhibitors). Additionally, we proposed an in vivo method for quantifying rates of autophagosome accumulation, which can be used to infer occurrence of autophagic flux. Last, we tested two FDA-approved drugs currently undergoing clinical studies for Parkinson disease, isradipine and nilotinib, and found that isradipine did not modulate autophagy, whereas nilotinib induced both autophagosome number and autophagic flux. It is hoped that others will find this line useful as an in vivo vertebrate model to find or validate autophagy modulators that might be used to halt the progression of neurodegenerative diseases. Abbreviations: 3MA: 3-methyladenine; BafA: bafilomycin A1; dd: dorsal diencephalon; dpf: days post fertilization; e: eye; eGFP: enhanced green fluorescent protein; Elavl3: ELAV like neuron-specific RNA binding protein 3; FDA: Food and Drug Administration; hb: habenula; hpt, hours post treatment; Map1lc3b: microtubule-associated protein 1 light chain 3 beta; nt: neural tube; ot, optic tectum; P/E: pepstatin A and E64d; PD: Parkinson disease; PMTs: photomultiplier tubes; PTU: 1-phenyl-2-thiourea; Ta: annealing temperature; Tel, telencephalon.