Maneuver-Based Trajectory Planning for Highly Autonomous Vehicles on Real Road With Traffic and Driver Interaction
DOI: 10.1109/tits.2010.2046037
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper presents a maneuver-based trajectory planning algorithm designed for highly autonomous vehicles operating on real roads with traffic and driver interaction. The research addresses the challenge of implementing trajectory planning on embedded systems with severe computational constraints, specifically targeting platforms with less than 150 MHz clock frequency, 150 KB RAM, and 3 MB program memory. Existing methods, such as tree exploration or potential field approaches, are deemed unsuitable due to high memory usage or computational costs. The work is part of the European HAVEit project, which aims to develop cooperative driving systems that allow for high-level automation while maintaining driver interaction and understanding. The proposed solution utilizes a two-step algorithm to balance computational efficiency with trajectory optimality. The first step, the maneuver module, identifies feasible maneuvers by combining three longitudinal actions (accelerate, decelerate, maintain speed) and three lateral actions (change left, change right, maintain lane), resulting in nine primary cases plus safe state and emergency maneuvers. This module performs a rapid risk assessment based on collision avoidance, using Equivalent Energetic Speed (EES) for crash severity and Time to Collision (TTC) and Inter-Vehicular Time (TIV) for collision probability. The output is a ranked list of maneuvers. The second step, the trajectory module, generates and evaluates specific trajectories within the accepted maneuvers. It optimizes for multiple criteria including safety, comfort, fuel consumption, speed, and traffic rules, while respecting vehicle dynamics constraints. The final output is a set of spatio-temporal points describing recommended vehicle states for the next 5 to 10 seconds. The system architecture supports different assistance modes, ranging from maneuver proposition to shared control, ensuring the automation’s decisions remain understandable to the driver. The risk assessment model calculates a composite risk value for each maneuver by combining the gravity of potential collisions with the probability of occurrence derived from traffic indicators. The trajectory generation process respects longitudinal and lateral acceleration limits to ensure controllability. The algorithm is designed to be interruptible to meet strict real-time requirements. The significance of this work lies in its ability to provide optimal, multi-criteria trajectory planning within the strict hardware limitations of commercial vehicle ECUs. By decoupling high-level maneuver selection from detailed trajectory optimization, the system achieves real-time performance without sacrificing safety or comfort. This approach facilitates a cooperative driving paradigm where the automation can intervene or assist the driver effectively, addressing the limitations of current Advanced Driver Assistance Systems (ADAS) that often lack deep integration with driver intent or sufficient computational efficiency for widespread commercial adoption. The paper validates the approach through simulation, demonstrating its feasibility for future highly automated driving systems.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | OpenAlex-citations | — | — | 1 | 2026-06-25 |
| archive | success | unpaywall | — | — | 2 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-25 |
| chunk | success | chunk | — | — | 1 | 2026-06-25 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-25 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-25 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Theoretical Contribution: computational model