Browse Prior Art Database

Method and system for auto-streamlining of driverless vehicle. Disclosure Number: IPCOM000241709D
Publication Date: 2015-May-26
Document File: 3 page(s) / 132K

Publishing Venue

The Prior Art Database


The invention teaches a multi-network architecture for controlling driverless vehicles and an algorithm which adjusts event activation such that a change of state decision is less probably for the driverless vehicle when in proximity to human driver driven vehicles. The multi-network architecture consists of a first server system corresponding to the controling agent where a driverless vehicle is currently located, a secondary server system corresponding to a target next location where said driverless vehicle is heading, and an ad hoc network distributed between priximate driverless vehicles. There is a hierarchial relationship between the three networks such that driverless vehicle change-solutions may be executed according to the system instructions w/priority or in some cases according to a voting scheme. In terms of the adjustable change-solution thresholds in the presence of a human driven vehicle, when the driverless vehicle identifies the priximity of a human driven vehicle, the network (centralized or distributed) adjusts proximate driverless vehicles change-solution thresholds such that a change of state for the driverless vehicle is less probably, thus reducing the over-all combined driverless vehicle variability or entropy/chaos in the presence of a human driven vehicle.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Method and system for auto-streamlining of driverless vehicle.

In any road, the traffic control and effective traffic management is totally depends on how effectively the vehicles are moving in the road. Improper streamlining of vehicles leads towards accommodating less number of vehicles in unit area of road. Following diagram (Figure 1) depicts vehicles not moving with proper streamlining. The vehicles are moving in haphazard way, so it is difficult to maintain the traffic and this may cause traffic jam and management of traffic.

Figure 1

Proposed is a method and system by which driverless vehicles can be arranges in a streamline automatically.

1. The centralized cloud server will gather the direction of movement of each vehicles and accordingly will align each vehicles in such a way that maximum number of vehicles can be accommodated in unit road area.

2. Software will consider the minimum clearance between any two vehicles while accommodating the maximum number of vehicles in unit read area.

3. Software will consider following parameters while auto-streamlining the vehicles in an optimum way.

a. Change in shape and dimension of the road across the road length

b. Target destination of each vehicles and current distance from the destination.

c. Dimension of each vehicles.

4. Based on the above information, software will auto-streamline the vehicles.

Software and controls could be built into each vehicle so the 'system' is distributed over all driverless cars in some locality. Changes in driving patterns for road conditions, emergency vehicles, weather conditions, etc. would also be necessary. In addition, logic for resolving "contentions" would also be required (e.g., slowing and speeding up of one car to allow


Page 02 of 3

another to enter the lane, etc.). A disadvantage of a centralized (e.g., cloud) solution is the

boundary problem--this could be handled by cloud-1 handing off a vehicle entering another cloud's service area (cloud-2), like with cell phones (but hand-offs are not perfect and a accident could happen during hand-off). The distributed system is superior since there would be no black-out times. FYI, the goal of streamlining a plurality of vehicles could be met in the distributed embodiment by each vehicle monitoring it's nearest neighbors position and velocity.

Another problem is the mixture of driverless and driver driven vehicles. Driver driven vehicles will be more erratic and variable and will complicate the achievement of the driverless vehicle cluster of streamlining. However, in a distributed system since each driverless vehicle is only interested in the positions and velocities of its nearest neighbors, the variabi...