Safe and Efficient Dynamic Window Approach for Differential Mobile Robots With Stochastic Dynamics Using Deterministic Sampling

We propose an efficient and safe dynamic window approach (DWA) by using deterministic sampling. When the system dynamics have uncertainty, the control input includes errors, so that the DWA objective function becomes a random variable. When a random-choice algorithm with a finite number of samples is used to estimate the objective function, it may miss collisions during prediction. In this work, we approximate the end-state distribution as a one-dimensional distribution for each input candidate in advance and generate sample paths deterministically to eliminate the misses to achieve safe control. Numerical experiments have shown that this method is approximately three times as efficient as the Monte Carlo method in most indoor environments.

• Publication year

  2023

• Venue

  IEEE Robotics and Automation Letters

• Publication date

  2023-05-01

• Fields of study

  Computer Science, Engineering

• Identifiers
  DOI 10.1109/LRA.2023.3257681
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Improvement of Dynamic Window Approach Using Reinforcement Learning in Dynamic Environments

  2022cited by this paper
• Tube Stochastic Optimal Control for Nonlinear Constrained Trajectory Optimization Problems

  2020cited by this paper
• DWA-RL: Dynamically Feasible Deep Reinforcement Learning Policy for Robot Navigation among Mobile Obstacles

  2020cited by this paper
• Path Planning Techniques for Mobile Robots A Review

  2020cited by this paper
• Adaptive Dynamic Window Approach for Local Navigation

  2020cited by this paper
• Predictive Collision Avoidance for the Dynamic Window Approach

  2019influential reference
• A Survey of Autonomous Driving: Common Practices and Emerging Technologies

  2019cited by this paper
• Motion Planning Among Dynamic, Decision-Making Agents with Deep Reinforcement Learning

  2018cited by this paper
• A Review of Motion Planning Techniques for Automated Vehicles

  2016cited by this paper
• Decentralized non-communicating multiagent collision avoidance with deep reinforcement learning

  2016cited by this paper
• Unscented guidance

  2015cited by this paper
• A comparison of Gaussian and Fourier methods for degenerate diffusions on SE(2)

  2015cited by this paper
• On sample size control in sample average approximations for solving smooth stochastic programs

  2013cited by this paper
• The Banana Distribution is Gaussian: A Localization Study with Exponential Coordinates

  2012cited by this paper
• Smooth and collision-free navigation for multiple robots under differential-drive constraints

  2010cited by this paper
• Information theory on Lie groups and mobile robotics applications

  2010cited by this paper
• Lectures on Stochastic Programming: Modeling and Theory

  2009cited by this paper
• Reciprocal Velocity Obstacles for real-time multi-agent navigation

  2008cited by this paper
• Iterative linearization methods for approximately optimal control and estimation of non-linear stochastic system

  2007cited by this paper
• Modified Newton's method applied to potential field-based navigation for mobile robots

  2006cited by this paper
• Probabilistic models of dead-reckoning error in nonholonomic mobile robots

  2003cited by this paper
• Motion Planning in Dynamic Environments Using Velocity Obstacles

  1998cited by this paper
• New extension of the Kalman filter to nonlinear systems

  1997cited by this paper
• Accurate odometry and error modelling for a mobile robot

  1997cited by this paper
• The dynamic window approach to collision avoidance

  1997cited by this paper
• Exact robot navigation using artificial potential functions

  1992influential reference

• A Safe Maritime Path Planning Fusion Algorithm for USVs Based on Reinforcement Learning A* and LSTM-Enhanced DWA

  2026cites this paper
• Digital Twin Modeling Method Based on the Dynamic Characteristics of Mobile Robots

  2025cites this paper
• Hybrid Path Planning Algorithm for Autonomous Mobile Robots: A Comprehensive Review

  2025cites this paper
• ALVO: Adaptive Learning with Velocity Obstacles for UGV Navigation in Dynamic Scenes

  2025cites this paper
• Nonholonomic Narrow Dead-End Escape with Deep Reinforcement Learning

  2025cites this paper
• Dynamic path planning for multiple unmanned ground vehicles based on improved dynamic window approach

  2025cites this paper
• A dynamically hybrid path planning and obstacle avoidance algorithm for mobile robots based on improved A-star and dynamic windows approach

  2025cites this paper
• AUV Collision Avoidance Strategy Based on Fuzzy Reinforcement Learning

  2025cites this paper
• DWAS-RL: A safety-efficiency balanced reinforcement learning approach for path planning of Unmanned Surface Vehicles in complex marine environments

  2025cites this paper
• Embodied Escaping: End-to-End Reinforcement Learning for Robot Navigation in Narrow Environment

  2025cites this paper
• A Comparative Study of Human Motion Models in Reinforcement Learning Algorithms for Social Robot Navigation

  2025cites this paper
• Gradient Field-Based Dynamic Window Approach for Collision Avoidance in Complex Environments

  2025cites this paper
• A Two-Stage Local Path Planning Algorithm Based on Sampling and Optimization Methods

  2025cites this paper
• Environment-Adaptive Motion Planning via Reinforcement Learning-Based Trajectory Optimization

  2025cites this paper
• A Path Planning Framework for Robots Based on Improved Parallel Sampling RRT and Offset Guidance DWA

  2025cites this paper
• ASAP: Agile and Safe Pursuit for Local Planning of Autonomous Mobile Robots

  2024cites this paper
• Multi-Robot Scheduling for Deadlock Avoidance Using Nonstop Areas

  2024cites this paper
• Probabilistic Representation and Its Application of Digital-Twin of Spatio-Temporal Real-World Toward Trustable Cyber-Physical Interactions

  2024cites this paper
• EPPE: An Efficient Progressive Policy Enhancement framework of deep reinforcement learning in path planning

  2024cites this paper
• An adaptive dynamic window approach for UUV obstacle avoidance planning in 3D environments

  2024cites this paper