English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Techniques based on Reinforcement Learning (RL) are increasingly being usedto design control policies for robotic systems. RL fundamentally relies onstate-based reward functions to encode desired behavior of the robot and badreward functions are prone to exploitation by the learning agent, leading tobehavior that is undesirable in the best case and critically dangerous in theworst. On the other hand, designing good reward functions for complex tasks isa challenging problem. In this paper, we propose expressing desired high-levelrobot behavior using a formal specification language known as Signal TemporalLogic (STL) as an alternative to reward/cost functions. We use STLspecifications in conjunction with model-based learning to design modelpredictive controllers that try to optimize the satisfaction of the STLspecification over a finite time horizon. The proposed algorithm is empiricallyevaluated on simulations of robotic system such as a pick-and-place roboticarm, and adaptive cruise control for autonomous vehicles.

Model-based Reinforcement Learning from Signal Temporal Logic Specifications