Enhancing the Generalization Performance and Speed Up Training for DRL-based Mapless Navigation

03/22/2021
by   Wei Zhang, et al.
9

Training an agent to navigate with DRL is data-hungry, which requires millions of training steps. Besides, the DRL agents performing well in training scenarios are found to perform poorly in some unseen real-world scenarios. In this paper, we discuss why the DRL agent fails in such unseen scenarios and find the representation of LiDAR readings is the key factor behind the agent's performance degradation. Moreover, we propose an easy, but efficient input pre-processing (IP) approach to accelerate training and enhance the performance of the DRL agent in such scenarios. The proposed IP functions can highlight the important short-distance values of laser scans and compress the range of less-important long-distance values. Extensive comparative experiments are carried out, and the experimental results demonstrate the high performance of the proposed IP approaches.

READ FULL TEXT

page 1

page 4

research
12/29/2021

DDPG car-following model with real-world human driving experience in CARLA

In the autonomous driving field, the fusion of human knowledge into Deep...
research
06/12/2023

High-speed Autonomous Racing using Trajectory-aided Deep Reinforcement Learning

The classical method of autonomous racing uses real-time localisation to...
research
02/18/2021

Causal Inference Q-Network: Toward Resilient Reinforcement Learning

Deep reinforcement learning (DRL) has demonstrated impressive performanc...
research
02/01/2022

Accelerating Deep Reinforcement Learning for Digital Twin Network Optimization with Evolutionary Strategies

The recent growth of emergent network applications (e.g., satellite netw...
research
02/15/2020

Map-less Navigation: A Single DRL-based Controller for Robots with Varied Dimensions

Deep reinforcement learning (DRL) has shown great potential in training ...
research
02/24/2023

Prioritized Trace Selection: Towards High-Performance DRL-based Network Controllers

Deep Reinforcement Learning (DRL) based controllers offer high performan...
research
11/21/2022

Disentangled Representation Learning

Disentangled Representation Learning (DRL) aims to learn a model capable...

Please sign up or login with your details

Forgot password? Click here to reset