Fleet Traffic Planning With Precomputed Resource Occupancy Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing centralized control methods for fleets of battery-electric vehicles struggle with frequent recharging intervals, longer recharging cycles, and limited processing power, necessitating efficient traffic planning that optimizes battery usage and minimizes standstill time while avoiding deadlock states and maintaining a high productivity-to-cost ratio.
Innovation Solution
A traffic planning method that generates a search tree with scores incorporating short-term and long-term components, using a memory to pre-compute long-term values from simulated or real vehicle movements, ensuring efficient vehicle control with moderate processing resources, and avoiding battery depletion risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If centralized control optimizes fleet productivity by coordinating vehicle movements, then productivity increases, but computational resources and processing power requirements increase
Solution Approach 1:
The patent segments the centralized control problem into two distinct components: an online traffic planner that handles real-time tactical decisions with limited computational resources, and an offline scorer that pre-computes long-term performance metrics for different resource occupancy states. This segmentation allows the online planner to operate efficiently with moderate processing power while the offline scorer performs intensive computations during idle periods to build lookup tables that guide future decisions.
Solution Approach 2:
The patent applies preliminary action by pre-computing score values for various resource occupancy configurations using simulated or historical vehicle movements. These pre-computed scores are stored in lookup tables that the online traffic planner can quickly query during real-time operation. This eliminates the need for complex real-time simulations and enables fast decision-making with limited computational resources.
2Productivity
If battery-electric vehicles operate close to totally discharged condition to maximize utilization, then battery usage efficiency improves, but risk of battery depletion and vehicle failure increases
Solution Approach 1:
The patent incorporates battery state feedback into the scoring mechanism. The scorer evaluates resource occupancy states not only based on traffic flow efficiency but also on vehicle battery levels and recharging requirements. This feedback loop ensures that the traffic planner receives comprehensive information about both operational and energy constraints when making routing decisions.
Solution Approach 2:
The patent applies beforehand cushioning by incorporating safety margins into the scoring function that penalize resource occupancy configurations leading to dangerous battery levels. The pre-computed scores inherently account for recharging needs and battery depletion risks, providing a cushion against failure by guiding vehicles away from occupancy patterns that would lead to critical battery states before they occur.
3Reliability
If traffic planning considers long-term resource occupancy patterns, then foresight and optimization improve, but computational complexity and processing time increase
Solution Approach 1:
The patent resolves this contradiction by performing the computationally intensive long-term pattern analysis in advance during an offline phase. The scorer pre-computes scores for various resource occupancy states by simulating vehicle movements and evaluating long-term performance metrics. These pre-computed scores are stored and can be quickly retrieved during online operation, providing foresight without real-time computational burden.
Solution Approach 2:
The patent segments the temporal horizon into online and offline components. The online traffic planner handles immediate tactical decisions with short-term horizon using pre-computed information, while the offline scorer handles long-term strategic analysis. This temporal segmentation allows long-term foresight to be incorporated without compromising real-time response speed.
4Measurement precision
If the traffic planner uses extensive simulation and evaluation of vehicle movements, then planning accuracy improves, but processing power requirements and computational cost increase
Solution Approach 1:
The patent applies preliminary action by performing extensive simulations and evaluations during an offline phase to build comprehensive score tables. The scorer uses simulated or historical vehicle movements to evaluate resource occupancy states with high accuracy, storing the results for quick retrieval. This shifts the computational burden from real-time operation to offline preparation, achieving high planning accuracy without real-time processing power constraints.
Solution Approach 2:
The patent uses copying by creating simplified representations of complex vehicle movement patterns through simulation. Instead of running full physics-based simulations in real-time, the system uses pre-computed score tables that capture the essential outcomes of extensive virtual experiments. This copying approach maintains planning accuracy while dramatically reducing real-time computational requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented traffic planning method (100) for controlling a plurality of vehicles which are movable among multiple shared resources, comprising: receiving (112) a transport mission; generating (114) a root node representing an initial resource occupancy of the vehicles; generating (116) a search tree from the root node, in which each edge represents a motion command and each node represents a resource occupancy, wherein each node is associated with a score related to the vehicles' fulfilment of the transport mission; identifying (118) a target node with an acceptable score; and deriving (120) a planned sequence of motion commands corresponding to a path to the target node, wherein the score of a node includes a short-term component, which represents a cost of executing all motion commands from the root node to said node, and a long-term component, which is determined by the resource occupancy that the node represents.