Bi-objective Motion Planning Approach for Safe Motions: Application to a Collaborative Robot

Abstract

This paper presents a new bi-objective safetyoriented path planning strategy for robotic manipulators. Integrated into a sampling-based algorithm, our approach can successfully enhance the task safety by guiding the expansion of the path towards the safest configurations. Our safety notion consists of avoiding dangerous situations, e.g. being very close to the obstacles, human awareness, e.g. being as much as possible in the human vision field, as well as ensuring human safety by being as far as possible from human with hierarchical priority between human body parts. Experimental validations are conducted in simulation and on the real Baxter research robot. They revealed the efficiency of the proposed method, mainly in the case of a collaborative robot sharing the workspace with humans.

Design of Advanced Human–Robot Collaborative Cells for Personalized Human–Robot Collaborations

“…User awareness is embedded in the path planner using a cost function that depends on the human state. The typical approach minimizes a weighted sum of an efficiency term (e.g., the path length) and user-aware terms, such as human-robot distance [43], trajectory repeatability [44], or human visibility [45].…”

Safety-Aware Time-Optimal Motion Planning With Uncertain Human State Estimation

“…Casalino et al [12] deform the trajectory according to a repulsion field associated with checkpoints on the human skeleton. Tarbouriech and Suleiman [19] propose a sampling-based algorithm that biases the search towards regions far from the current human state. Javdani et al [20] and Kanazawa et al [21] pose the problem of reducing the execution time of robot actions by avoiding idle times in handover tasks.…”

“…All the methods above realize an arbitrary compromise between some desired cost terms. For example, [19], [23] find a trade-off between the minimization of the path length and the maximization of the human-robot distance, while [21] minimizes a trade-off of the final error, joint bound satisfaction, and human-robot collision probability. However, this approach does not model the delays caused by safety stops and slowdowns when the human is close to the robot.…”

Motion Planning for Mobile Robots—Focusing on Deep Reinforcement Learning: A Systematic Review

“…At present, the DRL motion planning is mostly used to solve the problem of navigation and collision avoidance in the low dimensional plane ground environment, while the research on the comprehensive planning in the threedimensional space is less, such as the land marine amphibious robot and land air cooperative bionic robot system [139] . Compared with two-dimensional land space, high-dimensional motion planning will face more complex dynamic constraints and more uncertainties.…”