Mobile Robot Trajectory Planning for Dynamic Obstacle Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous robots face challenges in navigating environments effectively due to the inability to predict and avoid dynamic obstacles, leading to potential collisions and inefficient path planning.
Innovation Solution
A computer-implemented method using a predictive model, such as a deep neural network, to generate occupancy predictions for future time steps, allowing the robot to plan a trajectory that accounts for the movement of dynamic obstacles and minimizes collision risks by analyzing sensor data and motion models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the robot uses traditional path planning methods without predictive models, then the system complexity is low, but the robot cannot predict and avoid dynamic obstacles leading to collisions
Solution Approach 1:
The system performs preliminary actions by generating occupancy predictions for future time steps before the robot actually moves. The predictive model forecasts obstacle locations at future time steps, allowing the robot to plan its trajectory in advance while avoiding collisions with dynamic obstacles that would otherwise be unpredictable.
Solution Approach 2:
The patent introduces an intermediary predictive model that acts as a mediator between the robot's current state and future obstacle locations. This model generates occupancy predictions that bridge the gap between current sensor data and future collision risks, enabling the robot to make informed navigation decisions without directly observing future obstacle positions.
2Reliability
If the robot plans path avoiding all potential obstacle locations, then collision avoidance improves, but the path efficiency decreases due to overly conservative routing
Solution Approach 1:
The system applies local quality by generating occupancy predictions only for relevant future time steps and locations along the robot's potential trajectory. Rather than conservatively avoiding all possible obstacle locations globally, the predictive model focuses computational resources on predicting obstacles in specific local regions and time steps that are most relevant to the robot's near-term navigation decisions.
Solution Approach 2:
The patent implements partial action by generating occupancy predictions for a limited number of future time steps rather than attempting to predict all possible future obstacle locations indefinitely. This partial prediction approach provides sufficient information for safe navigation while avoiding the excessive computational burden of complete long-term prediction, thus maintaining path planning efficiency.
3Reliability
If the robot uses predictive models for obstacle location, then collision avoidance improves, but the computational requirements and processing time increase
Solution Approach 1:
The system applies partial action by generating occupancy predictions for a limited number of future time steps rather than attempting to predict all possible future obstacle locations indefinitely. This partial prediction approach provides sufficient information for safe navigation while avoiding the excessive computational burden of complete long-term prediction.
Solution Approach 2:
The patent implements continuous useful action by updating occupancy predictions continuously as the robot moves and new sensor data becomes available. Rather than performing discrete, computationally intensive replanning operations, the system maintains continuous predictive models that are updated incrementally, reducing processing time while maintaining prediction accuracy.
Data Source
AI summary
This specification describes trajectory planning for robotic devices. A robotic navigation system can obtain, for each of multiple time steps, data representing an environment of a robot at the time step. The system generates a series of occupancy maps for the multiple time steps, and uses the series of occupancy maps to determine occupancy predictions for one or more future time steps. Each occupancy prediction can identify predicted locations of obstacles in the environment of the robot at a different one of the future time steps. A planned trajectory can be determined for the robot using the occupancy predictions, and the robot initiates travel along the planned trajectory.