Deep reinforcement learning for map-less goal-driven robot navigation

Matej Dobrevski (Avtor), Danijel Skočaj (Avtor)

Povzetek

Mobile robots that operate in real-world environments need to be able to safely navigate their surroundings. Obstacle avoidance and path planning are crucial capabilities for achieving autonomy of such systems. However, for new or dynamic environments, navigation methods that rely on an explicit map of the environment can be impractical or even impossible to use. We present a new local navigation method for steering the robot to global goals without relying on an explicit map of the environment. The proposed navigation model is trained in a deep reinforcement learning framework based on Advantage Actor–Critic method and is able to directly translate robot observations to movement commands. We evaluate and compare the proposed navigation method with standard map-based approaches on several navigation scenarios in simulation and demonstrate that our method is able to navigate the robot also without the map or when the map gets corrupted, while the standard approaches fail. We also show that our method can be directly transferred to a real robot.

Ključne besede

mobilna robotika;spodbujevalno učenje;navigacija;globoko učenje;mobile robotics;reinforcement learning;navigation;deep learning;

Podatki

Jezik:	Angleški jezik
Leto izida:	2021
Tipologija:	1.01 - Izvirni znanstveni članek
Organizacija:	UL FRI - Fakulteta za računalništvo in informatiko
UDK:	007.52:004.8
COBISS:	60005379
ISSN:	1729-8814
Št. ogledov:	32
Št. prenosov:	8
Ocena:	0 (0 glasov)
Metapodatki:

Ostali podatki

Sekundarni jezik:	Slovenski jezik
Sekundarne ključne besede:	mobilna robotika;spodbujevalno učenje;navigacija;globoko učenje;
Vrsta dela (COBISS):	Članek v reviji
Strani:	str. 1-13
Letnik:	ǂVol. ǂ18
Zvezek:	ǂno. ǂ1
Čas izdaje:	Jan./Feb. 2021
DOI:	10.1177/1729881421992621
ID:	17908342