Anytime navigation with Progressive Hindsight optimization

In multi-robot systems, efficiently navigating in a a partially-known environment is an ubiquitous but challenging task, as each robot must account for the uncertainty introduced, for example, by other moving robots. This uncertainty makes pre-computed plans not always applicable, and often hinders the desired efficient use of the robot's resources. In this work, we present a local anytime approach for robot motion planning that accounts for the uncertainty of the environment by generating ‘snapshots’ of possible future scenarios. Our approach adapts the Hindsight optimization technique to allow robots to plan their immediate motion based on long-term efficiency. We validate our approach by comparing the efficiency on the paths executed against a state-of-the art navigation technique in a variety of scenarios, and show that by accounting for the uncertainty in the environment, agents can improve their time- and energy-efficient motions.

Authors

• Julio Godoy
• Ioannis Karamouzas
• Stephen J. Guy
• Maria L. Gini

Keywords

• Navigation
• Optimization
• Planning
• Uncertainty
• Collision avoidance
• Robot kinematics
• Hindsight Optimization
• Variety Of Scenarios
• Environmental Uncertainty
• Motion-based
• Use Of Robots
• Sequence Of Actions
• Planning Process
• Multiple Agents
• Reward Function
• Configuration Space
• Environmental Knowledge
• Emergent Behavior
• Environmental Agents
• Simulation Time Step
• Planning Time
• Performance Of Agents
• Simulated Cycle
• Goal Position
• Dynamic Obstacles
• Concept Of Optimization
• Velocity Of Agent
• Plan Generation