scholarly journals Online Mapping and Motion Planning Under Uncertainty for Safe Navigation in Unknown Environments

2021 ◽  
pp. 1-23
Author(s):  
Eric Pairet ◽  
Juan David Hernandez ◽  
Marc Carreras ◽  
Yvan Petillot ◽  
Morteza Lahijanian
Keyword(s):  
Motion Planning ◽  
Planning Under Uncertainty ◽  
Safe Navigation ◽  
Unknown Environments ◽  
Online Mapping

scholarly journals Synergistic exploration and navigation of mobile robots under pose uncertainty in unknown environments

2018 ◽  
Vol 15 (1) ◽  
pp. 172988141775078 ◽  
Author(s):  
Ioannis Arvanitakis ◽  
Anthony Tzes ◽  
Konstantinos Giannousakis
Keyword(s):  
Path Planning ◽  
Target Position ◽  
Limited Range ◽  
Field Of View ◽  
Planning Under Uncertainty ◽  
Safe Navigation ◽  
Navigation Function ◽  
Unknown Environments ◽  
Arduous Task ◽  
Planning Processes

Path planning under uncertainty in an unknown environment is an arduous task as the resulting map has inaccuracies and a safe path cannot always be found. A path planning method is proposed in unknown environments towards a known target position and under pose uncertainty. A limited range and limited field of view range sensor is considered and the robot pose can be inferred within certain bounds. Based on the sensor measurements a modified map is created to be used for the exploration and path planning processes, taking into account the uncertainty via the calculation of the guaranteed visibility and guaranteed sensed area, where safe navigation can be ensured regardless of the pose-error. A switching navigation function is used to initially explore the space towards the target position, and afterwards, when the target is discovered to navigate the robot towards it. Simulation results highlighting the efficiency of the proposed scheme are presented.

scholarly journals Deep Reinforcement Learning Approach with Multiple Experience Pools for UAV’s Autonomous Motion Planning in Complex Unknown Environments

Sensors ◽  
2020 ◽  
Vol 20 (7) ◽  
pp. 1890 ◽  
Author(s):  
Zijian Hu ◽  
Kaifang Wan ◽  
Xiaoguang Gao ◽  
Yiwei Zhai ◽  
Qianglong Wang
Keyword(s):  
Reinforcement Learning ◽  
Motion Planning ◽  
Predictive Control ◽  
Experimental Testing ◽  
The Novel ◽  
Simulation Environment ◽  
Unknown Environments ◽  
Autonomous Motion ◽  
Policy Gradient ◽  
Speed Up

Autonomous motion planning (AMP) of unmanned aerial vehicles (UAVs) is aimed at enabling a UAV to safely fly to the target without human intervention. Recently, several emerging deep reinforcement learning (DRL) methods have been employed to address the AMP problem in some simplified environments, and these methods have yielded good results. This paper proposes a multiple experience pools (MEPs) framework leveraging human expert experiences for DRL to speed up the learning process. Based on the deep deterministic policy gradient (DDPG) algorithm, a MEP–DDPG algorithm was designed using model predictive control and simulated annealing to generate expert experiences. On applying this algorithm to a complex unknown simulation environment constructed based on the parameters of the real UAV, the training experiment results showed that the novel DRL algorithm resulted in a performance improvement exceeding 20% as compared with the state-of-the-art DDPG. The results of the experimental testing indicate that UAVs trained using MEP–DDPG can stably complete a variety of tasks in complex, unknown environments.

NRR: a nonholonomic random replanner for navigation of car-like robots in unknown environments

Robotica ◽  
2014 ◽  
Vol 32 (7) ◽  
pp. 1101-1123 ◽  
Author(s):  
Ellips Masehian ◽  
Hossein Kakahaji
Keyword(s):  
Motion Planning ◽  
Mobile Robots ◽  
Random Tree ◽  
Local Motion ◽  
Goal Region ◽  
Unknown Environments ◽  
Effectiveness And Efficiency ◽  
Path Lengths

SUMMARYIn this paper, a new sensor-based approach called nonholonomic random replanner (NRR) is presented for motion planning of car-like mobile robots. The robot is incrementally directed toward its destination using a nonholonomic rapidly exploring random tree (RRT) algorithm. At each iteration, the robot's perceived map of the environment is updated using sensor readings and is used for local motion planning. If the goal was not visible to the robot, an approximate path toward the goal is calculated and the robot traces it to an extent within its sensor range. The robot updates its motion to goal through replanning. This procedure is repeated until the goal lies within the scope of the robot, after which it finds a more precise path by sampling in a tighter Goal Region for the nonholonomic RRT. Three main replanning strategies are proposed to decide when to perform a visibility scan and when to replan a new path. Those are named Basic, Deliberative and Greedy strategies, which yield different paths. The NRR was also modified for motion planning of Dubin's car-like robots. The proposed algorithm is probabilistically complete and its effectiveness and efficiency were tested by running several simulations and the resulting runtimes and path lengths were compared to the basic RRT method.

Sign in / Sign up

Export Citation Format

Share Document