Autonomous Vehicle Trajectory Planning with Safe Arrival Time Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches for autonomous vehicle path planning do not account for potential changes in the motion or state of objects in the environment, leading to safety concerns and increased computational requirements for frequent path updates.
Innovation Solution
The system generates paths and analyzes safety by considering potential changes in object motion, using sensor data to compute safe arrival times and visualize trajectories to ensure collision avoidance, allowing for longer paths and less frequent updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional heat map approaches are used for path planning, then collision probabilities are calculated based on object locations, but the paths become less safe over time and require frequent updates, increasing computational resource requirements
Solution Approach 1:
The system performs preliminary action by computing safe arrival times for all target positions in advance, before the vehicle actually reaches them. This allows the system to pre-determine which positions are safe to occupy at specific times, enabling longer path planning horizons without requiring frequent re-computations. The safe arrival time for each position is calculated based on predicted object trajectories and safety constraints, allowing the vehicle to plan ahead while ensuring collision avoidance.
Solution Approach 2:
The invention adds the time dimension to traditional spatial path planning by computing safe arrival times for each target position. Instead of only considering spatial coordinates (x, y), the system now considers (x, y, t) tuples, where t represents the safe arrival time. This transforms the path planning from a 2D spatial problem to a 3D space-time problem, enabling the vehicle to account for temporal aspects of object motion and create more reliable long-term paths.
2Productivity
If conventional approaches generate paths without accounting for object motion changes, then computational resources are reduced, but the paths may not provide sufficient safety assurance
Solution Approach 1:
The system changes parameters by computing safe arrival times that incorporate predicted object motion parameters such as velocity, acceleration, and trajectory. Instead of using static object locations, the system dynamically adjusts the safe arrival time calculations based on object motion parameters, allowing it to account for potential changes in object state while maintaining computational efficiency through optimized algorithms.
3Reliability
If frequent path updates are performed to maintain safety, then path safety is improved, but significant processing power, energy, and bandwidth are required
Solution Approach 1:
By computing safe arrival times in advance for all relevant target positions, the system eliminates the need for frequent path re-planning. The pre-computed safe arrival time information allows the vehicle to make safe navigation decisions throughout its journey without requiring continuous re-computation, thereby significantly reducing energy consumption while maintaining high safety standards.
Data Source
AI summary
In various examples, sensor data representative of a field of view of at least one sensor of a vehicle in an environment is received from the at least one sensor. Based at least in part on the sensor data, parameters of an object located in the environment are determined. Trajectories of the object are modeled toward target positions based at least in part on the parameters of the object. From the trajectories, safe time intervals (and/or safe arrival times) over which the vehicle occupying the plurality of target positions would not result in a collision with the object are computed. Based at least in part on the safe time intervals (and/or safe arrival times) and a position of the vehicle in the environment a trajectory for the vehicle may be generated and/or analyzed.