Vehicle Trajectory Prediction for Collision-Aware Path Planning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Autonomous and semi-autonomous vehicle planning systems face challenges in reliably assessing the reliability of sensor data for complex scenarios, leading to potential inaccuracies in determining vehicle actions to avoid obstacles.

Innovation Solution

Implementing a model on a vehicle computing device to predict object trajectories and vehicle interactions, using sensor data, map data, and control policies to determine potential intersections and optimize computational resources for accurate and efficient path planning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If planning systems utilize sensor data to determine vehicle actions in complex scenarios, then the ability to avoid objects is improved, but the complexity of the system increases and reliability inspection becomes difficult

Engineering Contradiction:

Improvereliability of sensor data assessmentVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the complex planning task into distinct modules: a perception module that generates object trajectories from sensor data, and a planning module that receives these trajectories as input. This segmentation allows each module to be independently validated and tested, improving reliability inspection despite overall system complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces object trajectories as an intermediary representation between sensor data and planning decisions. These trajectories serve as a standardized interface that decouples the perception and planning modules, allowing reliability assessment of sensor data processing without requiring inspection of the entire complex system.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If the system processes multiple object trajectories and potential interactions, then the accuracy of predicting vehicle-object intersections is improved, but the computational resources required increase

Engineering Contradiction:

Improveaccuracy of intersection predictionVSAvoidcomputational resource consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary processing by generating object trajectories in advance using the perception module. By pre-computing these trajectories and filtering out implausible ones, the planning module receives refined input that requires less computational processing to accurately predict intersections, thereby reducing overall computational resource consumption.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12441308B2Active prediction based on object trajectories

Publication Date: 2025.10.14 ZOOX INC

AI summary

Techniques for accurately predicting and avoiding collisions with objects detected in an environment of a vehicle are discussed herein. A vehicle computing device can implement a model to output data indicating costs for potential intersection points between the object and the vehicle in the future. The model may employ a control policy and a time-step integrator to determine whether an object may intersect with the vehicle, in which case the techniques may include predicting vehicle actions by the vehicle computing device to control the vehicle.