Unified path planner for parking an autonomous vehicle based on RRT

Abstract

Maneuvering autonomous vehicles in constrained environments, such as autonomous vehicle parking, is not a trivial task and has received increasing attention from both the academy and industry. However, the traditional methods divide the problem into parallel parking, perpendicular parking, and echelon parking, then different methods are applied for the parking motion planning. In this paper a Rapidly-exploring Random Tree (RRT) based path planner is implemented for autonomous vehicle parking problem, which treats all the situations in a unified manner. As the RRT method sometimes generates some complicated paths, a smoother is also implemented for smoothing generated paths. The proposed algorithm is verified in simulation and generates applicable solutions for the proposed application scenarios.

A unified motion planning method for parking an autonomous vehicle in the presence of irregularly placed obstacles

“…To briefly summarize here, compared to the prevailing methods that can only deal with specific cases (e.g., [13,15,16,18,20,21,26,27]), our formulated model describes various kinds of cases in a unified way regardless they are regular parking scenarios (i.e., parallel, perpendicular or echelon parking cases) or irregular ones.…”

“…First, many existing methods do not solve the motion control problem directly. Typically, those heuristic-based path planning methods suffer from this limitation because kinematic descriptions of the vehicle are either missing or incomplete (e.g., [15,16,[19][20][21]). In fact, quite few works have formulated complete kinematics (e.g., [27]).…”

“…Here, a path refers to a geometric curve y ¼ f ðxÞ in the xy coordinate frame, whereas a trajectory attaches the time course along a path, i.e., the determination of x ¼ xðtÞ [17]. Heuristic-based methods usually seek solutions from artificial intelligence techniques, e.g., fuzzy logics [18,19], search-based methods [20,21], random sampling methods [22] and machine learning methods [23].…”

Vehicle path planning with maximizing safe margin for driving using Lagrange multipliers

“…Their approach generates good results in practice for urban driving. Other common methods in autonomous driving literature, such as Rapidly-exploring Random Trees (RRTs) or Probabilistic Roadmap (PRM), generate nodes randomly to construct paths which satisfy vehicle kinematic or dynamic requirements though they do not consider path clearance [2], [6], [7]. In an algorithm based on the RRT to develop a path planner [8].…”

A Practical and Optimal Path Planning for Autonomous Parking Using Fast Marching Algorithm and Support Vector Machine

“…Other planners [13]- [15] have been successfully dealing with vehicle's dynamic constraints by using probabilistic path planning methods. In [14] L. Han proposed a unified autonomous parking planner based on bi-directional RRT but this method does not consider the number of gear transition. Kuwata [15] developed algorithm of closed loop rapidly-exploring random tree (CL-RRT) for the path planning task running in real time.…”