Control of Heavy Duty Vehicle

Vehicle Safety Assessment and Safety Enhancement
Vehicle Safety

1. ENHANCED AHS SAFETY THROUGH THE INTEGRATION OF VEHICLE CONTROL AND COMMUNICATION

Task Order 4210

Karl Hedrick, University of California, Berkeley
khedrick@me.berkeley.edu, http://www.me.berkeley.edu/faculty/hedrick/

This project is a continuation of MOU388. It continues to develop three related concepts which exploit the cooperative nature of AHS (with vehicles communicating and coordinating with each other and the roadway) to yield safety and capacity gains. The first task uses communication systems to implement dynamic position tracking of vehicles on an AHS and fully coordinated platoon maneuvers. The second task will develop and experimentally test an algorithm that exploits the position tracking and communication abilities to estimate the friction characteristics of the road and construct a map of the roadway's friction characteristics as a function of location. The third task merges the results in the previous two tasks with existing and emerging PATH work on emergency maneuvers to produce a detailed simulation of emergency stopping of a platoon on slippery roads.

2. SAFETY PERFORMANCE AND ROBUSTNESS OF HEAVY DUTY VEHICLES

Task Order 4211

Christian Gerdes, Stanford University
gerdes@cdr.stanford.edu, http://www-me.stanford.edu/faculty/facultydir/gerdes.html

This project is a continuation of MOU390. It concludes ongoing research to make robust claims about the performance of heavy trucks with AVCSS technologies in safety-critical conditions, with the consideration of the complexity of truck dynamics that has made it difficult to establish what constitutes a safe truck. The project begins with the construction of a set of metrics to quantity safety measures for heavy trucks. The set of metrics will cover the safety demands (such as yaw stability, roll-over avoidance and stopping distance) of heavy trucks. Robustness of the safety claims will then be established by considering different operating conditions experienced by heavy trucks, variations in vehicle dynamic parameters arising through loading conditions, component choice and wear. Using a dynamic model, sensitivity analysis will be performed to determine which parametric variations produce the most significant impact on the safety metrics.

3. DETECTION AND AVOIDANCE OF COLLISIONS: THE ACT MODEL

Task Order 4220

George Anderson, University of California, Riverside
andersen@ucrac1.ucr.edu, http://www.psych.ucr.edu/faculty/Andersen.htm

This project investigates the visual information (e.g. angle of collision) used by human drivers to detect and avoid collisions. The research will involve conducting experiments, using a driving simulator, to derive the relevant model parameters of a general collision detection model referred to as the Assessment of Collision Threat (ACT) model. Based on the results of the experiments, microsimulations of the ACT model will be performed to evaluate the viability of the model to detect collisions in the car following and lane changing scenarios. The ACT model will be integrated with the human driver model being developed at PATH. Validation of the simulation based results will be performed by comparing driving performance from simulation studies with data from extant studies on real-world driving behavior

4. SAFETY ASSESSMENT OF ADVANCED VEHICLE CONTROL AND SAFETY SYSTEMS (AVCSS): A CASE STUDY

Task Order 4225

Wei-Bin Zhang, California PATH
wbzhang@path.berkeley.edu, http://www.path.berkeley.edu/PATH/General/Staff/wzhang.html

Ching-Yao Chan, California PATH
cychan@uclink4.berkeley.edu, http://www.path.berkeley.edu/PATH/General/Staff/cchan.html

This project is a continuation of MOU395. It continues the collaboration between PATH and INRETS in France in performing safety analysis and evaluation of advanced vehicle control and safety systems (AVCSS) such as vehicle longitudinal and lateral control systems, so that errors in the process of specification, design, development and integration can be revealed prior to the implementation of the new technologies and hazardous consequences can be prevented. PATH will contribute expertise in vehicle control and failure analysis to identify, analyse and classify the risks inherent in each part of an AVCSS system. INRETS will provide expertise in safety analysis and assessment. The project will also consider sensor reliability and diagnosis, and develop a system architecture to include fault tolerance and fault detection. As a case study, a transit bus will be used as a study platform.

5. EMERGENCY OPERATION OF PLATOONS: COLLISIONS, EMERGENCY DECELERATION AND PLATOON LANE CHANGE

MOU 319

Nancy Ann Lynch, Massachusetts Institute of Technology
lynch@theory.lcs.mit.edu, http://theory.lcs.mit.edu/~lynch/

Darbha Swaroop, Texas A&M
dswaroop@usha.tamu.edu, http://people.tamu.edu/~swaroop/

Hariharan Krishnan, California PATH
krishnan@path.berkeley.edu,

This project proposes to carry out a detailed design and analysis of regulation and low level coordination controllers for emergency platoon operation. The central theme of the work will be the issue of collisions within a platoon: we will incorporate collision modeling into our system description, determine the effect of collisions on the system performance (lateral deviation, passenger injury, vehicle damage) and attempt to optimize the system performance with respect to these considerations. We will also study the problem of obstacle avoidance and platoon lane change, and analyze the stability of different lane change schemes and establish necessary communication requirements.

Online papers:
PWP-99-3.pdf (411 K), PRR-98-36.pdf (685 K), PRR-99-5.pdf (4.1 mb), PRR-2000-1.pdf (411 K), PRR-2000-2.pdf (548 K)

6. OPTIMIZED VEHICLE CONTROL/COMMUNICATION INTERACTION IN AN AUTOMATED HIGHWAY SYSTEMS

MOU 320

Karl Hedrick, University of California, Berkeley
khedrick@me.berkeley.edu,

This goal of the project is to define and optimize the interaction between the communication system and the vehicle control system, from both a hardware and software standpoint. The project proposes to use existing vehicle control algorithms developed by PATH researchers at U.C. Berkeley and off-the-shelf communication technology to design a coordination protocol by these the control and communication systems can affect the various maneuvers required by an automated highway system.