Motion Planning in Complex Environments Using Closed-loop Prediction

Abstract

This paper describes the motion planning and control subsystems of Team MIT's entry in the 2007 DARPA Grand Challenge. The novelty is in the use of closed-loop prediction in the framework of Rapidly-exploring Random Tree (RRT). Unlike the standard RRT, an input to the controller is sampled, followed by the forward simulation using the vehicle model and the controller to compute the predicted trajectory. This enables the planner to generate smooth trajectories much more efficiently, while the randomization allows the planner to explore cluttered environment. The controller consists of a Proportional-Integral speed controller and a nonlinear pure-pursuit steering controller, which are used both in execution and in the simulation-based prediction. The main advantages of the forward simulation are that it can easily incorporate any nonlinear control law and nonlinear vehicle dynamics, and the resulting trajectory is dynamically feasible. By using a stabilizing controller, it can handle vehicles with unstable dynamics. Several results obtained using MIT's race vehicle demonstrate these features of the approach.

Vision-based guidance and control of a hovering vehicle in unknown, GPS-denied environments

“…Thus, trajectory generation is decoupled from control and estimation performance; such coupling would otherwise cause undesirable feedback. The same methodology was successfully employed by MIT in the DARPA Urban Challenge [25]. Fig.…”

Sampling-Based Threat Assessment Algorithms for Intersection Collisions Involving Errant Drivers

“…The initial velocity is v 1 (0) = 1.5m/s for the errant vehicle and v 2 (0) = 2.0m/s for the host vehicle, whose velocity is assumed constant. The RRTReach algorithm uses a tree size of 2000 nodes with a time horizon of T h = 3s, and relies on a pure-pursuit controller (Kuwata et al, 2008) to control the steering motion of the errant vehicle in its propagation step. To make the problem even more constrained, a small obstacle is added to the minor road around location (x,y) = (2.5, 0.5).…”

CF-Pursuit: A Pursuit Method with a Clothoid Fitting and a Fuzzy Controller for Autonomous Vehicles

“…Another improvement relates to the look-ahead distance. The look-ahead distance will change with the velocity, the command velocity [15]. The specific tuning mechanism can be obtained using the following expression:…”

“…MIT proposed a simplified adaptive method that tuned the look-ahead distance with the velocity. The tuning method's parameters were obtained from a number of different experiences [15,18].…”