Heuristic-free Optimization of Force-Controlled Robot Search Strategies in Stochastic Environments

07/15/2022
by   Benjamin Alt, et al.
0

In both industrial and service domains, a central benefit of the use of robots is their ability to quickly and reliably execute repetitive tasks. However, even relatively simple peg-in-hole tasks are typically subject to stochastic variations, requiring search motions to find relevant features such as holes. While search improves robustness, it comes at the cost of increased runtime: More exhaustive search will maximize the probability of successfully executing a given task, but will significantly delay any downstream tasks. This trade-off is typically resolved by human experts according to simple heuristics, which are rarely optimal. This paper introduces an automatic, data-driven and heuristic-free approach to optimize robot search strategies. By training a neural model of the search strategy on a large set of simulated stochastic environments, conditioning it on few real-world examples and inverting the model, we can infer search strategies which adapt to the time-variant characteristics of the underlying probability distributions, while requiring very few real-world measurements. We evaluate our approach on two different industrial robots in the context of spiral and probe search for THT electronics assembly.

READ FULL TEXT

page 1

page 5

research
08/05/2020

Search@Home: A Commercial Off-the-Shelf Environment for Investigating Optimization Problems

Search heuristics, particularly those that are evaluation-driven (e.g., ...
research
07/12/2023

Learning Hierarchical Interactive Multi-Object Search for Mobile Manipulation

Existing object-search approaches enable robots to search through free p...
research
03/14/2022

Synergistic Scheduling of Learning and Allocation of Tasks in Human-Robot Teams

We consider the problem of completing a set of n tasks with a human-robo...
research
08/24/2020

Feature Guided Search for Creative Problem Solving Through Tool Construction

Robots in the real world should be able to adapt to unforeseen circumsta...
research
01/23/2019

A Constraint Programming Approach to Simultaneous Task Allocation and Motion Scheduling for Industrial Dual-Arm Manipulation Tasks

Modern lightweight dual-arm robots bring the physical capabilities to qu...
research
04/21/2021

Learning and Planning for Temporally Extended Tasks in Unknown Environments

We propose a novel planning technique for satisfying tasks specified in ...
research
07/28/2016

MIST: Missing Person Intelligence Synthesis Toolkit

Each day, approximately 500 missing persons cases occur that go unsolved...

Please sign up or login with your details

Forgot password? Click here to reset