Automatic configuration of ROS applications for near-optimal performance

Abstract

The performance of a ROS application is a function of the individual performance of its constituent nodes. Since ROS nodes are typically configurable (parameterised), the specific parameter values adopted will determine the level of performance generated. In addition, ROS applications may be distributed across multiple computation devices, thus providing different options for node allocation. We address two configuration problems that the typical ROS user is confronted with: i) Determining parameter values and node allocations for maximising performance; ii) Determining node allocations for minimising hardware resources that can guarantee the desired performance. We formalise these problems with a mathematical model, a constrained form of a multiple-choice multiple knapsack problem. We propose a greedy algorithm for optimising each problem, using linear regression for predicting the performance of an individual ROS node over a continuum set of parameter combinations. We evaluate the algorithms through simulation and we validate them in a real ROS scenario, showing that the expected performance levels only deviate from the real measurements by an average of 2.5%.

Interactive Robot Transition Repair With SMT

“…If a precise model of the dependency between parameters and behaviors is available, it may be possible to design a calibration procedure that executes a specific sequence of actions and to recover correct parameter values (e.g., [1]). If a calibration procedure cannot be designed, but the effect of parameters is well-understood, it may be possible to optimize for the parameters using a functional model [5]. Model-based diagnosis can diagnose faulty parameters [7] if the behavior of the robot in its environment can be formally defined.…”

“…Configuration parameters are common in many software systems, and represent a large source of error across disciplines [3,4,5]. In robotics, behaviors often rely on environment-dependent parameters for robust and accurate execution.…”

Tuning Perception and Motion Planning Parameters for Moveit! Framework

“…The performance of the planners was benchmarked in a series of simulated motion planning problems using two industrial manipulators; however, no cameras were used to monitor the surrounding of the robots during the tests, thus the effects of the perception settings were not investigated. Cano et al [Cano 2016] in their work addresses the problem of automatic optimal configuration of distributed ROS system in terms of CPU utilisation and data flow from each of the optimised ROS nodes. The researchers first investigated the correlation between the setting of one critical parameter of each ROS node, its performance and CPU utilisation.…”

Automatic Parameter Tuning of Motion Planning Algorithms

“…In [10] it is proposed a benchmarking infrastructure to compare the performance of different planners, however nothing is said about tuning the individual parameters of each algorithm. Finally, [11], [12], [13] address the configuration problem within a distributed system where several processes/algorithms can be tuned with only one parameter. However, the parameter values can be affected by the specific allocations of processes to processors within the system.…”