Trajectory Optimization in Robotic Applications, Survey of Recent Developments

Abstract

Trajectory Optimization (TO) is the sequence of processes that are considered in order to produce the best path that mends the overall performance or reduces the consumption of the resources where the restriction system remains maintained. In this survey, an inclusive review of the latest advancements in modeling and optimization of trajectory generation in robotic applications will be discussed broadly. In recent times, numerous studies have employed optimal control techniques involving direct and indirect methods in order to convert the authentic Trajectory Optimization problem into a constrained parameter optimization problem. Moreover, a huge variety of optimization algorithms such as Genetic Algorithms (GA), Simulated Annealing (SA), Sequential Quadratic Programming (SQP), and Particle Swarm Optimization (PSO) are used aiming to find the optimal solutions for trajectory planning. It is observable that, minimizing the jerk, energy consumption, and execution time among the most widely used design objectives, on the other hand, the robot’s joints configurations and the motor torques are the most used design variables. This paper aims to review the fundamental techniques and their coincident robotic applications in the field of trajectory optimization aiming to afford some steering for related researchers.

An Integrated Programmable CPG with Bounded Output

“…These methods can be categorized into four groups, namely 1) time scaling methods deforming the original timing law of the desired trajectory to provide environment adaptation and/or safe motion [9]. 2) trajectory optimization methods design the optimum trajectory that mends the overall performance or reduces the consumption of the resources where the restricted system remains maintained [10]. 3) state space optimization methods search in the state space of the system for finding the optimum trajectory from the current state to the desired one [11], [12].…”