Receding Horizon Planning with Rule Hierarchies for Autonomous Vehicles

12/06/2022
by   Sushant Veer, et al.
0

Autonomous vehicles must often contend with conflicting planning requirements, e.g., safety and comfort could be at odds with each other if avoiding a collision calls for slamming the brakes. To resolve such conflicts, assigning importance ranking to rules (i.e., imposing a rule hierarchy) has been proposed, which, in turn, induces rankings on trajectories based on the importance of the rules they satisfy. On one hand, imposing rule hierarchies can enhance interpretability, but introduce combinatorial complexity to planning; while on the other hand, differentiable reward structures can be leveraged by modern gradient-based optimization tools, but are less interpretable and unintuitive to tune. In this paper, we present an approach to equivalently express rule hierarchies as differentiable reward structures amenable to modern gradient-based optimizers, thereby, achieving the best of both worlds. We achieve this by formulating rank-preserving reward functions that are monotonic in the rank of the trajectories induced by the rule hierarchy; i.e., higher ranked trajectories receive higher reward. Equipped with a rule hierarchy and its corresponding rank-preserving reward function, we develop a two-stage planner that can efficiently resolve conflicting planning requirements. We demonstrate that our approach can generate motion plans in  7-10 Hz for various challenging road navigation and intersection negotiation scenarios.

READ FULL TEXT

page 1

page 6

research
05/01/2021

Pedestrian Collision Avoidance for Autonomous Vehicles at Unsignalized Intersection Using Deep Q-Network

Prior research has extensively explored Autonomous Vehicle (AV) navigati...
research
10/08/2019

Multi-Vehicle Interaction Scenarios Generation with Interpretable Traffic Primitives and Gaussian Process Regression

Generating multi-vehicle interaction scenarios can benefit motion planni...
research
09/29/2020

Modeling and Testing Multi-Agent Traffic Rules within Interactive Behavior Planning

Autonomous vehicles need to abide by the same rules that humans follow. ...
research
11/09/2020

Behavior Planning at Urban Intersections through Hierarchical Reinforcement Learning

For autonomous vehicles, effective behavior planning is crucial to ensur...
research
06/20/2018

A Look at Motion Planning for Autonomous Vehicles at an Intersection

Autonomous vehicles are currently being tested in a variety of scenarios...
research
01/27/2023

Tree-structured Policy Planning with Learned Behavior Models

Autonomous vehicles (AVs) need to reason about the multimodal behavior o...
research
09/16/2023

GRaCE: Optimizing Grasps to Satisfy Ranked Criteria in Complex Scenarios

This paper addresses the multi-faceted problem of robot grasping, where ...

Please sign up or login with your details

Forgot password? Click here to reset