GB2579021A

GB2579021A - Vehicle control system and method

Info

Publication number: GB2579021A
Application number: GB1818526.4A
Authority: GB
Inventors: Richard Allsopp Samuel; Andrew Jardine Fraser; Anwar Mahmoud Taie Mostafa
Original assignee: Jaguar Land Rover Ltd
Current assignee: Jaguar Land Rover Ltd
Priority date: 2018-11-14
Filing date: 2018-11-14
Publication date: 2020-06-10
Anticipated expiration: 2038-11-14
Also published as: GB201818526D0; GB2579021B

Abstract

There is disclosed a control system and a method for a host vehicle operable in an autonomous mode. The control system comprises one or more controllers. The control determines a traffic condition, beyond a first sensing range of the host vehicle and associates the traffic condition with a first lane of a multi-lane highway comprising a plurality of lanes in a first direction. If there is traffic associated with the lane the host vehicle is travelling in, the lane the host vehicle is in is changed, before the host vehicle reaches the location of the traffic, to another of the plurality of lanes of the multi-lane highway. The lane change may be performed at a greater time and/or distance from the traffic condition than a lane change associated with a vehicle overtaking function of the host vehicle. If the traffic is associated with a junction that is intended to be used, the vehicle may prioritise the first lane if a lane change is to be made.

Description

VEHICLE CONTROL SYSTEM AND METHOD

TECHNICAL FIELD

The present disclosure relates to a vehicle control system and method. In particular, but not exclusively it relates to a vehicle control system and method for a vehicle operable in an autonomous mode.

Aspects of the invention relate to a control system, a method, a vehicle, computer software, and a non-transitory computer-readable storage medium.

BACKGROUND

It is known for a vehicle to host a system that enables the host vehicle to operate in accordance with a predefined autonomous mode. The host vehicle may be instructed to operate in accordance with the predefined autonomous mode by a user (occupant) of the host vehicle i.e. via an input device at which a user input is received to control operation of the predefined autonomous mode.

The occupant may desire for the speed and path of the host vehicle in the autonomous mode to be appropriate to a driving context. The driving context may relate to factors such as the environment outside the host vehicle. The environment includes infrastructure and other road users (ORUs). The driving context may relate to the specific preferences of the occupant. The driving context may relate to the condition of the host vehicle.

It is an aim of the present invention to address disadvantages of the prior art.

SUMMARY OF THE INVENTION

Aspects and embodiments of the invention provide a control system, a method, a vehicle, computer software, and a non-transitory computer-readable storage medium.

According to an aspect of the invention there is provided a control system for a host vehicle operable in an autonomous mode, the control system comprising one or more controllers, the control system configured to: determine a traffic condition, beyond a first sensing range of the host vehicle; associate the traffic condition with a first lane of a multi-lane highway comprising a plurality of lanes in a first direction including the first lane; and cause control of a direction of the host vehicle before the host vehicle reaches the location of the traffic condition to change lane from one of the plurality of lanes to another of the plurality of lanes of the multi-lane highway, in dependence on the traffic condition being associated with the first lane. This advantageously reduces the chance of the host vehicle being disrupted or boxed in by other traffic moving out of the first lane.

The one or more controllers may collectively comprise: at least one electronic processor having an electrical input for receiving the information; and at least one electronic memory device electrically coupled to the at least one electronic processor and having instructions stored therein; and wherein the at least one electronic processor is configured to access the at least one memory device and execute the instructions thereon so as to cause the host vehicle to perform the determining, the associating and the causing the control.

The lane change may be performed at a greater time and/or distance before the host vehicle reaches the traffic condition than a lane change associated with a vehicle overtaking function of the host vehicle. This advantageously reduces the chance of the host vehicle being boxed in by other traffic.

The lane change may be farther before the host vehicle reaches the traffic condition for changing from an offside lane to a nearside lane, compared to changing from a nearside lane to an offside lane. This advantageously reduces the chance of the host vehicle being boxed in by other traffic while remaining compliant with highway law.

The control may be performed in dependence on confirmation of the traffic condition using information from one or more sensors of the host vehicle having the first sensing range, and/or from communication with another road user proximal to the traffic condition. This verification advantageously reduces the impact of a false positive determination.

The traffic condition may comprise a slower-moving or stationary traffic queue in the first lane. The method advantageously reduces the chance of being stuck in traffic.

The traffic condition may be associated with a junction and wherein the control system may be configured to determine in dependence on navigation information whether the host vehicle is to exit the multi-lane highway at the junction, wherein if the host vehicle is to exit the multi-lane highway at the junction and the one lane is a second lane different from the first lane, the another lane is the first lane. This advantageously reduces the chance of traffic blocking access to a junction.

The traffic condition may comprise a closure of at least one of the plurality of lanes, wherein the associating comprises determining that the closure is of at least the first lane, and wherein if the one lane is the first lane, the another lane is a second lane different from the first lane. This advantageously reduces the chance of unnecessarily waiting in traffic for the first lane to clear until the closure is visible and it is apparent that a lane change is needed.

The determining and/or the associating may utilise dynamic information received from a remote information source. The determining and/or the associating may utilise at least one of: map data comprising information on junctions and/or lane closures; road sign information comprising information on junctions; dynamic map data comprising information on lane closures; or dynamic traffic data comprising information on traffic conditions. This advantageously enables traffic conditions to be anticipated before other road users are aware of them.

The control system may be configured to select the another lane from the plurality of lanes, in dependence on one or more of: a determination that traffic speed in the another lane is faster than traffic speed in a different one or more of the plurality of lanes; or a navigation constraint. This advantageously enables travel using the fastest available lane.

The control of the direction may comprise causing a steering subsystem of the host vehicle to control steering of the host vehicle to follow a planned path of the host vehicle determined in dependence on the determining.

According to another aspect of the invention there is provided a method for controlling a host vehicle operable in an autonomous mode, the method comprising: determining a traffic condition, beyond a first sensing range of the host vehicle; associating the traffic condition with a first lane of a multi-lane highway comprising a plurality of lanes in a first direction including the first lane; and causing control of a direction of the host vehicle before the host vehicle reaches the location of the traffic condition to change lane from one of the plurality of lanes to another of the plurality of lanes of the multi-lane highway, in dependence on the traffic condition being associated with the first lane.

According to a further aspect of the invention there is provided a vehicle comprising any one or more of the control systems described herein.

According to a further aspect of the invention there is provided computer software that, when executed, is arranged to perform any one or more of the methods as described herein.

According to a further aspect of the invention there is provided a non-transitory, computer-readable storage medium storing instructions thereon that, when executed by one or more electronic processors, causes the one or more electronic processors to carry out any one or more of the methods as described herein.

Within the scope of this application it is expressly intended that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, in the claims and/or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. That is, all embodiments and/or features of any embodiment can be combined in any way and/or combination, unless such features are incompatible. The applicant reserves the right to change any originally filed claim or file any new claim accordingly, including the right to amend any originally filed claim to depend from and/or incorporate any feature of any other claim although not originally claimed in that manner.

BRIEF DESCRIPTION OF THE DRAWINGS

One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Fig 1 illustrates an example of a host vehicle; Fig 2A illustrates an example of an electronic controller; Fig 2B illustrates an example of a computer-readable storage medium; Fig 2C illustrates an example of a system; Fig 3 illustrates an example of a host vehicle on a highway; Fig 4 illustrates an example of a method; Fig 5 illustrates an example of a method; Fig 6 illustrates an example trajectory of a host vehicle; and Fig 7 illustrates an example of a method.

DETAILED DESCRIPTION

Fig 1 illustrates an example of a vehicle 10 in which embodiments of the invention can be implemented. In some, but not necessarily all examples, the (host) vehicle 10 is a passenger vehicle, also referred to as a passenger car or as an automobile. Passenger vehicles generally have kerb weights of less than 5000 kg. In other examples, embodiments of the invention can be implemented for other applications, such as industrial vehicles.

The host vehicle 10 may comprise any appropriate prime mover (not shown) or a plurality of prime movers. An example of a prime mover is an internal combustion engine.

Another example of a prime mover is an electric motor. The vehicle may be an electric vehicle or a hybrid-electric vehicle.

The host vehicle 10 may be operable in an autonomous mode. The host vehicle 10 may also be operable in a non-autonomous mode.

A control system 200 is shown in Fig 2A which may implement, at least in part, the functionality of the autonomous mode. The control system 200 may implement, at least in part, the functionality of the non-autonomous mode. The control system 200 may comprise means to cause any one or more of the methods described herein to be performed, at least in part.

The control system 200 may comprise one or more (electronic) controllers 202. One controller 202 is shown in Fig 2A.

The controller 202 of Fig 2A includes at least one electronic processor 204; and at least one electronic memory device 206 electrically coupled to the electronic processor 204 and having instructions 208 (e.g. a computer program) stored therein, the at least one electronic memory device 206 and the instructions 208 configured to, with the at least one electronic processor 204, cause any one or more of the methods described herein to be performed.

The control system 200 may be supplied separately from or together with any input devices and any actuators controlled by the control system 200.

Fig 2B illustrates a non-transitory computer-readable storage medium 210 comprising the computer program 208 (computer software).

Fig 20 shows an example of a system 300 for a vehicle such as the host vehicle 10 of Fig 1. The system 300 may implement, at least in part, the functionality of the autonomous mode.

The system 300 comprises the control system 200. The system 300 may comprise one or more actuators for operation by at least the control system 200 in at least the autonomous mode. The actuators may be operably (directly or indirectly) coupled to one or more outputs of one or more controllers of the control system 200.

The actuators may comprise one or more torque control actuators. The torque control actuators are for controlling torque received at one or more road wheels of the host vehicle 10.

The torque control actuators may comprise a brake control actuator 316.

The brake control actuator 316 comprises any appropriate means for controlling a negative torque received by road wheels of the host vehicle 10.

In an implementation, the brake control actuator 316 may comprise a friction brake actuator for applying friction brakes of the host vehicle 10.

The brake control actuator 316 may be operated in dependence on an output signal such as a brake demand signal output from the control system 200, in the autonomous mode.

The torque control actuators may comprise an acceleration control actuator 312.

The acceleration control actuator 312 comprises any appropriate means for controlling a positive torque received by road wheels of the host vehicle 10, for instance means for controlling a torque output of the prime mover.

In an implementation, the acceleration control actuator 312 may comprise a throttle position actuator for controlling an opening degree of a throttle for an internal combustion engine.

The acceleration control actuator 312 may be operated in dependence on an output signal such as a torque demand signal output from the control system 200, in the autonomous mode.

The actuators may comprise a steering control actuator 314. The steering control actuator is part of a steering subsystem of the host vehicle 10, for controlling the direction of the host vehicle 10.

The steering control actuator 314 comprises any appropriate means for controlling a direction of the host vehicle 10, for instance means for controlling a steering angle of front road wheels of the host vehicle 10.

In an implementation, the steering control actuator 314 may comprise a motor for actuating a steering rack of the host vehicle 10. Additionally or alternatively, the steering control actuator 314 may comprise a friction brake actuator, configured to control a braking torque differential between left and right wheels of the host vehicle 10.

The steering control actuator 314 may be operated in dependence on an output signal such as a steering signal output from the control system 200, in the autonomous mode.

One or more of the actuators 312, 314, 316 may be operable automatically by the control system 200 in the autonomous mode. One or more of the actuators may be operable under manual control by a vehicle occupant in the non-autonomous mode.

The system 300 may comprise one or more input devices 304, 306, 308, 310. The input devices may be coupled to one or more inputs of one or more controllers of the control system 200.

The signals to the actuators may be dependent on signals from the input devices.

The input devices may comprise sensing means 306, 308, 310, such as one or more sensor units. The sensing means may enable machine vision for autonomous driving.

The sensing means outputs to the control system 200 environment information indicative of the environment in the vicinity of the host vehicle 10. The environment information is indicative of one or more environment characteristics, e.g. road type, presence of other road users, road markings, road priorities, etc. The sensing means may be configured for or suitable for attachment to the host vehicle 10.

The sensing means may comprise an optical sensor such as a (visual) camera 308. An optical sensor is for imaging in the visible light spectrum.

The sensing means may comprise range detection means 310. The term "range detection means" will be understood to mean any sensing means for detecting sensor data indicative of a range of a target object from the host vehicle 10. The range detection means 310 may comprise a rangefinder. The range detection means 310 may comprise a laser rangefinder. The laser rangefinder may comprise a lidar sensor. The control system 200 with at least one of the sensing means may be arranged to capture a doppler shift in an emitted signal. The sensing means may comprise a radar sensor 306. The sensing means may comprise an ultrasound sensor (not shown).

The system 300 may comprise a plurality of the input devices, each input device representing a different sensing modality. For example, the system 300 may comprise lidar sensors, radar sensors and cameras. This redundancy improves safety and enables autonomous driving in various environments such as driving at night or through fog.

The sensing means may be capable of detecting objects within a first sensing range. The first sensing range may be, at most, a maximum line of sight distance from the sensing means. The first sensing range may be from approximately 80m to approximately 100m from the location of the sensing means.

Objects may be recognised by a classification process (algorithm) of the control system 200.

Objects which may be classified may include one or more of: automobiles; heavy goods vehicles; motorcycles or pushbikes; emergency services vehicles; road signs and their instructions (including temporary street furniture such as traffic cones); and road markings and their instructions. The locations of the objects may be determined, for example using range detection means 310. Which lane the objects are in may be determined. A relative speed between the host vehicle 10 and an object may be determined, which may indicate whether a separation (also referred to as headway, inter-vehicular distance, following distance) from the object is increasing or decreasing and at what rate. The movement of the objects may be tracked using optical flow analysis for example.

The sensing means collectively provide a field of view around the host vehicle 10. The field of view may extend 360 degrees horizontally around the host vehicle 10 or less. The collective field of view also extends vertically by any appropriate amount. The individual sensor units may be located at the front, rear and/or sides of the host vehicle 10. Sensor units may be located at corners of the host vehicle 10. Sensor units may be located on wing mirrors of the host vehicle 10. Some sensor units may be located high on the host vehicle 10, such as above the waist of the host vehicle 10. Some sensor units may be at bumper height or lower.

The input devices may communicate with the control system 200 using any appropriate electronic communication network. Similarly, the actuators may be configured for drive-by-wire operation, therefore communication between the control system 200 and the actuators may also be via any appropriate electronic communication network. Redundancy may be provided by implementing multiple communication networks and/or backup controllers in the control system 200 and/or backup power supplies coupled to independent power sources (e.g. batteries). Example communication networks include a Controller Area Network (CAN), an Ethernet network, a Local Interconnect Network, a FlexRay(TM) network or the like.

The system 300 may comprise a telematics unit 304. The telematics unit 304 may comprise one or more controllers. The telematics unit 304 may be a telematics control unit (TCU). The illustrated TCU does not form part of the control system 200 but it may do in other examples. The TCU may be configured at least to function as a vehicle software update client. The TCU may comprise an antenna arrangement. The antenna arrangement may be configured as a receiver, a transmitter or as a transceiver. This enables software updates to be obtained from a remote (offboard) information source 302 such as a server, another vehicle according to a vehicle-to-vehicle (V2V) communication model, or external infrastructure according to a vehicle-to-infrastructure (V2 I) communication model.

The TCU may be configured to download software-over-the-air (SOTA) updates for installation in the host vehicle 10. Software components for SOTA updates could include at least one of: executable code, configuration data, graphics, map data, dynamic data such as dynamic map data and dynamic traffic data and weather data, audio calibration, multimedia, and firmware.

SOTA updates are received via a wireless communication network, such as a cellular network. The host vehicle 10 may subscribe to a cellular network service. The TCU may comprise a subscriber identity such as an international mobile subscriber identity (IMSI) number, to facilitate access to the cellular network. A subscriber identity module (SIM) may be installed in the host vehicle 10 to enable the TCU to access the IMSI and therefore the cellular network. An operator of the cellular network may associate the IMSI with a customer account and bill the customer for data usage and/or access to the cellular network. Additionally or alternatively, the TCU may comprise means to access a short-range communication network such as a wireless local area network or a wireless personal area network. The TCU may comprise means, such as a universal serial bus interface, for wired communication with the remote information source 302.

Advantageously, SOTA functionality enables the dynamic data to be downloaded while the host vehicle 10 is undergoing a journey. This enables substantially live updates. The telematics unit 304 may be configured to schedule the SOTA downloads from the remote information source 302 according to push or pull methods. Client-server, V2V and/or V2I communication models could be used. The telematics unit 304 may be configured to perform the downloads periodically at a predetermined interval which may depend on the download payload. For instance, the interval for downloading dynamic traffic data may range from the order of minutes to the order of hours. The interval for downloading dynamic map data may range from the order of minutes to the order of months. The interval for downloading non-dynamic data may range from the order of months to the order of years, or it may need to be manually updated at a dealership.

Dynamic traffic data as described above may be obtained via a SOTA update and/or a service provider application programming interface. Dynamic traffic data comprises substantially live information on traffic conditions. For example, the dynamic traffic data may indicate slow moving or stopped traffic. The dynamic traffic data may be associated with one or more metrics associated with traffic density, flow rate, speed, inter-vehicular distance, or journey time. The metrics may indicate a current condition, a change, or an expected condition. The metrics may be associated with particular locations and/or with particular times. Falling speeds/flow rates/inter-vehicular distances and rising densities/journey times are indicators of traffic conditions.

The dynamic traffic data enables a traffic condition to be determined. The traffic condition could be determined by comparing a current condition with a change or expected condition. A traffic condition could be determined when at least one threshold is passed such as an absolute or relative threshold. The relative threshold could be a statistical significance threshold, for example.

The dynamic traffic data may have sufficient resolution, granularity and/or detail to enable the traffic condition to be associated with a specific lane of a highway, from a plurality of lanes for travel in a same direction. This enables certain lanes to be avoided before a traffic queue is reached.

Dynamic map data as described above may comprise information that enables map data stored onboard the host vehicle 10 to be supplemented. The map data may be used by the control system 200 and/or a navigation subsystem of the host vehicle 10 for route planning. Map data indicates at least roads and junctions. Locations may be indicated by map data via global position coordinates. The navigation subsystem may be configured to receive user navigation inputs defining navigation constraints. Navigation constraints may comprise one or more of a destination, a waypoint, a navigation route or acceptable routes, an avoidance setting (e.g. avoidance of toll roads), a target to be reduced/minimised such as minimum distance or minimum travel time, or a target to be achieved such as a time of departure or arrival. Once a navigation route has been selected, the selected navigation route may impose navigation constraints on the autonomous mode, to enable autonomous navigation.

The dynamic map data and dynamic traffic data may be compatible with said map data.

Dynamic map data may comprise indications of at least one of the following conditions: traffic conditions such as roadworks and/or lane closures; speed limit changes such as variable speed limit changes imposed by permanent variable speed limit systems; weather conditions; or road surface conditions. Examples of roadworks include road closures, lane closures and traffic diversion routes. Examples of lane closures include blocked lanes, whether caused by roadworks, broken down vehicles or other causes. Examples of road surface conditions include potholes, loose or broken surface material, low friction hazards (e.g. ice or spilled liquids), or objects in the road (e.g. lost cargo). The indications may specify one or more locations such as where the condition starts and/or ends. The indications may specify which lane or lanes the condition applies to. The indications enable certain lanes or roads to be avoided before a traffic queue is reached. The above indications may be available by analysis of data from the sensing means, however for a much shorter range. Indications from multiple sources, such as the dynamic map data and the sensing means, may be combined to improve certainty.

The map data, dynamic map data and/or dynamic traffic data may comprise a fine level of granularity. For example, the individual lanes for travel in a same direction on a highway may be distinguishable. The map data and/or dynamic map data may comprise a high level of detail. For example, indications of road markings and/or road sign (traffic sign) information may be distinguishable from the data. Distinguishable road markings may comprise indications of lane boundaries. Distinguishable lane boundaries may be indicated by lane boundary markings in the data or may be indirectly indicated by lane centre position information in the data. The map data and/or dynamic map data may be of any suitable format that enables an identification of an instruction regarding a lane, a junction, a right of way (priority) or caution.

The control system 200 may further be configured to determine a highway law applicable to the host vehicle's current location and/or to a planned location or route of the host vehicle 10. The control system 200 may be configured to apply information associated with the applicable highway law to correctly identify instructions from the map data and/or dynamic map data. For example, if a planned route is in the United Kingdom the control system 200 may be configured to recognize road markings or traffic (road) sign information in a manner that corresponds to the requirements of the Highway Code. This is advantageous because the same road markings or signs can have different legal meanings in different highway jurisdictions.

The additional detail from the map data and/or dynamic map data may enable not only improved route planning by a navigation subsystem, but also improved path planning for the autonomous mode. For example, the control system 200 may process the map data and/or dynamic map data to determine which lanes the host vehicle 10 will travel on at which points on a journey. The control system 200 may further determine when lane changes may need to occur as directed by road signs or other information from the data. Certain lanes can be avoided or moved out of before a traffic queue is reached. Further, the dynamic data may define a second sensing range of machine vision, farther than the first sensing range. For example, the dynamic data may at least cover an entire route planned by the navigation subsystem and may cover one or more alternative routes in case of a later route recalculation. This enables certain lanes or roads to be avoided. The dynamic data may cover a regional, national or even international area. However, a greater coverage area may adversely affect a time taken to download updated dynamic data.

The input devices may define one or more sensing modes for detecting objects or contexts such as road markings, road signs or traffic conditions, etc. The map data/dynamic data may define a further sensing mode for detecting at least some of the same objects or contexts. Therefore, some objects and contexts can be determined from plural modes of information.

The control system 200 may be configured to aggregate the multi-modal information and process the aggregated data to increase a confidence score of at least one property of the object or context. The property may relate to a presence or absence of the object or context, its location, its size, or anything else useful for autonomous mode driving. This advantageously enables a realistic indication of a driving context within at least the first sensing range. A required manoeuvre may only be performed if the confidence score is above a threshold.

A decision to perform a manoeuvre may be made on the basis of information from a longer-range low-trust sensing mode such as map data and/or dynamic data, but it may be required that the information leading to the decision is subsequently verified using a shorter-range high-trust sensing mode such as the sensing means, for the manoeuvre to be performed. For example, information from the sensing means may be used to verify that information from the map data/dynamic data is accurate, before one or more planned manoeuvres are performed. The longer-range low-trust sensing mode may correspond to map data and/or dynamic data. The shorter-range high-trust sensing mode may correspond to one or more of the above-described sensing means.

Other dynamic data that may be obtainable by the control system 200, e.g. via the TCU, may include dynamic traffic data indicative of an emergency services vehicle location. The dynamic traffic data may indicate if an emergency services vehicle is approaching. This provides advance warning for the host vehicle 10 to manoeuvre out of a position in which it would obstruct the emergency services vehicle. The data may be received from client-server, V2V and/or V2I communication.

The host vehicle 10 may additionally comprise at least one human-machine interface (HMI) (not shown), facilitating access to one or more of the functions of the control system 200 described herein, and/or for presenting one or more outputs of the control system 200 described herein to the occupant (e.g. driver). The presentation may use visual means, audio means or any other appropriate means. User inputs to the HMI may be via touch, gesture or sound-based commands, or any other appropriate means. The HMI may comprise one or more of an output HMI, an input HMI, or an input-output HMI. Examples of output HMI in a vehicle include a centre console display, an instrument cluster display, audio speakers, a head-up display, a rear seat occupant display, a haptic feedback device, or the like. Examples of input HMI include touchscreens, manual actuators such as buttons and switchgear, and sensors for speech command recognition or non-touch gesture recognition. The input HMI may be close to a driver's seat. Advantageously, some input HMI may be located on the steering wheel.

A handover process may be implemented for initiating the autonomous mode, which will now be described. The control system 200 may be configured to receive at least one signal indicative of a suitability of initiation of the autonomous mode. The received signal may be indicative of a vehicle characteristic. The received signal may be indicative of a user characteristic. The received signal may be indicative of an environment characteristic. The received signal may be from the sensing means, or from another part of the control system 200 such as an algorithm that processes the map data and/or dynamic data.

The control system 200 may be configured to cause output of an availability signal indicative of an availability of the autonomous mode in dependence on the received signal, for example for presentation to the occupant via an HMI. If no availability signal is output, the host vehicle 10 is not operable to enter the autonomous mode. The control system 200 may be configured to determine whether to transmit the availability signal in dependence on at least one of the vehicle characteristic, the user characteristic, or the environment characteristic. One or more criteria associated with one or more of the characteristics may need to be satisfied, for the availability signal to be transmitted. An indication of the availability signal may be continuously presented to the occupant until at least one of the criteria is no longer satisfied. The availability signal may be continuously presented to the occupant until a user input is received in response to the availability signal. Examples of the user input and examples of the criteria are defined below.

The control system 200 may be configured to receive the user input in the form of a user activation signal indicative of the occupant's request to initiate the autonomous mode in response to the availability signal. The user input may be made via HMI. The user activation signal may be received during driving of the host vehicle 10, in other words while the host vehicle 10 is in a travelable state. For example, the host vehicle 10 may be in the non-autonomous mode. The control system 200 may be configured to output a driving mode signal to cause the host vehicle 10 to initiate the autonomous mode in response to the user activation signal. Initiating the autonomous mode may comprise a transition phase during which control of vehicle movement is transitioned away from the occupant to the control system 200. A duration of the transition phase may be dependent on one or more of the vehicle characteristic, the user characteristic or the environment characteristic to ensure a smooth transition.

The environment characteristic may be indicative of an environment external to the host vehicle 10 and in the vicinity of the host vehicle 10. The environment may be a driving environment. The driving environment may be a current driving environment while the host vehicle 10 is being driven. The driving environment may be indicative of a type of road on which the host vehicle 10 is driving. Optionally, the control system 200 may be configured not to transmit the availability signal unless at least the environment characteristic satisfies a road type criterion. The environment characteristic may be indicative of other environments too.

The road type criterion may be satisfied if the environment characteristic is indicative that the host vehicle 10 is travelling on a required type of road. The required type may be a motorway. Articles 1(j) and 25 of the Vienna convention on road traffic define the term motorway. A motorway may be referred to as a freeway or highway in some countries. The term 'highway' is used in this document. For those countries which have not ratified the above convention, their highways are defined herein as those which possess many or all of the following characteristics of a highway: - Use of the road is prohibited to pedestrians, animals, cycles, mopeds unless they are treated as motor cycles, and all vehicles other than motor vehicles and their trailers, and to motor vehicles or motor-vehicle trailers which are incapable, by virtue of their design, of attaining on a flat road a speed specified by domestic legislation. This indication may be provided by a road sign; - Drivers are forbidden to have their vehicles standing or parked elsewhere than at marked parking sites; if a vehicle is compelled to stop, its driver shall endeavour to move it off the carriageway and also off the flush verge and, if he is unable to do so, immediately signal the presence of the vehicle at a distance so as to warn approaching drivers in time; - Drivers are forbidden to make U-turns, to travel in reverse, and to drive on to the central dividing strip, including the crossovers linking the two carriageways; - Drivers emerging on to a motorway shall give way to vehicles travelling on it; -The road is specially signposted as a motorway; - The road does not cross at level with any road, railway or tramway track, or footpath; - The road does not serve properties bordering on it; - The road is provided, except at special points or temporarily, with separate carriageways for the two directions of traffic, separated from each other either by a dividing strip not intended for traffic.

The road type criterion may not be satisfied if the road is of another type and/or does not possess all or at least certain ones of the above characteristics. For example, some roads are main roads that possess many of the above characteristics but allow pedestrians and 16 non-motorized vehicles to use the roads. The availability signal may not be transmitted for such roads.

In other examples, the required type of road may be another type of road rather than a highway, or the requirement may merely be that the host vehicle 10 is not on a certain type of road such as a minor or urban road. Optionally, the road may be required to possess multiple lanes in a direction of travel of the host vehicle 10 to satisfy the road type criterion. In other examples, there may be no road type criterion for entering the autonomous mode.

The driving environment such as the type of road may be determined directly from metadata in the map data. The metadata may be indicative that the road is a highway. Alternatively, the required type may be determined indirectly from indications that the road possesses one or more of the above characteristics. Indications of the above characteristics may be determined by recognition of relevant road signs or road markings conveying these requirements, or by recognition of infrastructure such as a dividing strip. This may be detected by the sensing means and recognized by an object classification algorithm or determined from the map data or dynamic map data.

The environment characteristic may be indicative of a current weather condition in the vicinity of the host vehicle 10 or an upcoming weather condition to be encountered by the host vehicle 10. Information indicative of a weather condition may be indicative of rain falling on the host vehicle 10. The information may be indicative of the presence of snow or ice on the ground. The information may be indicative of at least one of a temperature, a humidity, a wind speed, a visibility, atmospheric pressure, precipitation. The control system 200 may be configured to not output the availability signal unless at least one weather criterion is satisfied. weather criterion may be satisfied if an indicated weather condition is a predetermined acceptable weather condition or is not a predetermined unacceptable weather condition. A weather criterion may be satisfied if an indicated temperature is within a predetermined acceptable temperature range. A weather criterion may be satisfied if an indicated humidity is within a predetermined acceptable humidity range. A weather criterion may be satisfied if an indicated atmospheric pressure is within a predetermined acceptable pressure range. The weather condition may be determined from a sensor on the host vehicle 10 or from information downloaded from an offboard weather service.

The user characteristic may be indicative of a current user characteristic of the occupant of the host vehicle 10 while the host vehicle 10 is being driven by the occupant. The user characteristic may be indicative of an awareness of the occupant of the vehicle. Information indicative of the awareness of the occupant may be obtained from one or more user sensors (not shown). The one or more user sensors may comprise at least one of a camera 308 and a physiological sensor to capture data indicative of the awareness of the occupant. The control system 200 may be configured not to output the availability signal unless at least one awareness criterion is satisfied. The occupant's awareness may need to be above a predetermined awareness threshold to satisfy the awareness criterion. In an implementation, the awareness may be quantified by numerical indicators such as a frequency or length of time for which the occupant's gaze has not been within a predefined area associated with driving, a blink rate, a head pose angle, or the like. In other words, the autonomous mode may be unavailable to the occupant of the host vehicle 10 if the occupant is not sufficiently aware to be able to resume control of the host vehicle 10 from the autonomous mode if required. In some examples, the occupant characteristic may relate to a physiological state.

To satisfy a physiological criterion for the availability signal, quantifiable indicators such as heart rate or brain activity may be detected using one or more biometric sensors.

The user characteristic may be indicative of a separation of at least a part of the occupant from one or more controls of the host vehicle 10. For example, the user characteristic may be indicative of whether one or more hands of the occupant are on the steering wheel. The availability signal may be determined not to be output unless at least a non-separation criterion is satisfied. The non-separation criterion may be satisfied if one or more hands of the occupant are on the steering wheel.

The vehicle characteristic may be indicative of a current vehicle characteristic of the host vehicle 10 while the host vehicle 10 is being driven. The vehicle characteristic may be indicative of a current speed of the host vehicle 10. Information indicative of the current speed could be obtained from a speed sensor (not shown). The availability signal may be determined not to be output unless at least a speed criterion is satisfied. The speed criterion may be satisfied if an indicated current speed of the host vehicle 10 is within a predetermined acceptable speed range, such as less than an upper limit of about 130 kilometres per hour. Other vehicle characteristics may be checked too such that the availability signal is determined not to be output in one or more of the following situations: a tyre pressure is outside a predetermined acceptable range; an oil level is below a predetermined threshold; a fuel level is below a predetermined threshold; the host vehicle 10 is towing; a loaded weight of the host vehicle 10 exceeds a predetermined threshold; or a state of health of one or more components of the host vehicle 10, e.g. a traction battery, is outside a predetermined acceptable state of health.

The vehicle characteristic may be indicative of a detection range of one or more of the sensing means. The detection range may be less than the first sensing range in certain conditions, particularly weather conditions such as fog. The availability signal may be determined not to be output unless at least a detection range criterion is satisfied. The detection range criterion may be satisfied if the received signal is indicative that the detection range of the one or more sensing means is greater than a predetermined range threshold. The autonomous mode may be unavailable to the occupant of the host vehicle 10 if the detection range of the one or more sensors does not meet the predetermined range threshold.

Once the transition phase is entered, control of the host vehicle 10 moves away from the occupant and to the control system 200 of the host vehicle 10. The transition phase may comprise modifying a vehicle movement in preparation for the end of the transition phase.

For example, a steering of the host vehicle 10 may be controlled autonomously during the transition phase to substantially centre the host vehicle 10 within a lane of the road. A braking torque of the host vehicle 10 may be controlled autonomously during the transition phase to control a distance of the host vehicle 10 from a further vehicle ahead of the host vehicle 10 along a road. During the transition phase, the host vehicle 10 may also continue to respond to manual control inputs from the occupant. As the transition phase progresses, the host vehicle 10 may become less responsive to user control until the host vehicle 10 is controlled fully autonomously in the autonomous mode. The occupant is informed of progress through the transition phase by the transition signal described hereinbefore.

Once the transition phase is complete, the control system 200 controls the host vehicle 10 in the autonomous mode. SAE International's J3016 defines six levels of driving automation for on-road vehicles. The term autonomous mode as used herein will be understood to cover any of the SAE levels three or higher, such that the control system 200 will control all aspects of the dynamic driving task. At levels four or five, one or more aspects of one or more of the handover processes described herein for transitioning to and/or from the autonomous mode may not be implemented.

Driver-assistance functions such as cruise control, adaptive cruise control,a lane change assistance function, or a lane keeping function, are at a lower level of autonomy than the autonomous mode.

In the autonomous mode the occupant may not be required to keep one or more hands on the steering wheel, so a monitoring step requiring the occupant to keep one or more hands on the steering wheel may be omitted. In other implementations, the autonomous mode may require the monitoring step. Whether the hand(s) are on the steering wheel may be determined using any appropriate sensing means such as a touch sensor or camera or steering wheel torque/angle sensor. The monitoring may be performed periodically or continuously. If the hands are not on the steering wheel, one or more prompts may be issued.

The host vehicle 10 may comprise a driver distraction function. One or more distraction criteria associated with the driver distraction function may be inhibited upon entering the autonomous mode. For example, in the non-autonomous mode the driver distraction function may alert the occupant when their gaze points outside a predetermined area such as the windscreen. The alert may be transmitted when the gaze is outside the predetermined area for a threshold duration and/or frequency. In the autonomous mode the driver distraction function may be disabled or the predetermined thresholds may be modified to become more permissive.

While the host vehicle 10 is in the autonomous mode, one or more algorithms are implemented for controlling speed and/or direction of the host vehicle 10. The control system 200 transmits the output signals to the actuators in dependence on the algorithms. The algorithms may comprise at least some of: a lane centring algorithm; a lane change algorithm; a path planning algorithm; a speed control algorithm; a machine learning algorithm. The algorithms may be context-aware. The algorithms may process information from one or more of the sensing means: map data; dynamic data; and navigation constraints. For example, the algorithms may be traffic-aware from the dynamic traffic data.

The algorithms may interoperate with each other to determine the output signals. The algorithms may plan variations of the output signals over a future period of driving.

Algorithms for autonomous driving are known and include regression algorithms, classification algorithms, clustering, and decision matrix algorithms. Cost or loss functions may be employed to find optimal paths and speeds and minimize risk to humans.

The lane centring algorithm is for keeping the host vehicle 10 within a predetermined lateral position (target lane position) within lane lateral edges (lane boundaries). The lane boundaries may be identified by specific road markings under the relevant highway law. If road markings are not visible, for instance due to faded paint, a putative lane and/or its boundaries may be identified based on detection of a traffic corridor of other road users driving in a detected consistent manner, e.g. in lines.

The lane position may be off-centre on occasion, dependent on detected characteristics such as environment characteristics, e.g. other road users or infrastructure proximal to a lane boundary. This provides a reassuring separation between the host vehicle 10 and lateral objects. A minimum separation from one or both lane boundaries may be maintained. The minimum separation may be around 0.3 to 0.6 metres from the nearside boundary, optionally 0.5 metres.

The lane change algorithm may be for keeping the host vehicle 10 within a nearside lane if required by applicable highway law. The lane change algorithm may enable the host vehicle 10 to manoeuvre from a first lane to a second lane to avoid detected traffic. The lane change algorithm may enable the host vehicle 10 to implement a vehicle overtaking function to overtake another road user. The lane change algorithm may enable the host vehicle 10 to change lanes to follow a navigation route. A turn signal/indicator of the host vehicle 10 may be flashed automatically just before the lane change is performed.

Keeping the host vehicle 10 within a nearside lane may be the responsibility of a nearside bias function of the lane change algorithm. The nearside bias function may require a nearside lane to be selected in normal driving conditions. The nearside bias function may comprise one or more parameters that define constraints to be met. The constraints may be for lane hogging avoidance. An example constraint may be to delay changing lane from a nearside lane to an offside lane to overtake other road users until the overtake can be performed within a threshold time. A related constraint may be to change lane from the offside lane back to the nearside lane following an overtake as soon as possible. The threshold time may be the time spent outside the nearside lane without overtaking another road user in the nearside lane. The threshold may depend on applicable highway law but tends to be of the order of tens of seconds rather than minutes.

Whether a lane change is performed may depend on a space availability signal indicative of a presence of a space in front of or behind another road user of a size sufficient to accommodate the host vehicle 10, should the host vehicle 10 need to change lanes to occupy that space. The space availability signal may be determined in dependence on inputs from the sensing means. The space availability signal may affect where, when and/or how fast a manoeuvre is performed. For example, the space availability signal may be used by the speed control algorithm when the lane change algorithm determines a requirement for a lane change. The space may be in a target lane for the lane change. The space may be between a lead (downstream) other road user and a rear (upstream) other road user. The space may be a current or expected space. The control system 200 may be configured to determine if the expected space will have a size sufficient to accommodate the host vehicle 10 at a predetermined time in the future. Determination of the expected space may depend on a detected indication of a relative speed of the other road user or users. The speed may be controlled in dependence on the space availability signal, for example to ensure that the space in front of and behind the host vehicle 10 is of a sufficient, e.g. above-threshold, detected size. The threshold size is an example of a manoeuvring constraint to be satisfied before the manoeuvre can be performed. The threshold may depend on the speed of the host vehicle 10. The speed of the host vehicle 10 may be controlled before the lane change.

The speed may be controlled to be close to a speed of a lead other road user, a speed of a rear other road user, or between both.

The path planning algorithm may be for planning a specific path to be followed. Planning the path comprises determining one or more manoeuvre requirements indicative of required manoeuvres of the host vehicle 10. A manoeuvre is defined herein as a change of speed or course. Changing course may be performed using the steering control actuator.

Absent of navigation constraints, the path may follow the highway as far as possible. With navigation constraints, the path may follow those portions of the navigation route during which the autonomous mode is on or available. The path may extend beyond the first sensing range. The portion of the path within the first sensing range may be optimised.

Examples of optimisations include reducing/minimizing targets such as derivatives of velocity (acceleration, jerk) when steering the host vehicle 10. Cost functions or the like may be used to perform optimisations.

The speed control algorithm is for planning a required speed of the host vehicle 10 to be followed using the torque control actuators. The speed control algorithm may enable functions such as adaptive cruise control, overtaking speed boost, and lane changes. The speed control algorithm may also be for complying with a speed limit detected using road sign recognition or map data. The speed may be controlled in advance of traffic conditions beyond the first sensing range, indicated for example by the dynamic data. The speed control algorithm may determine a speed to maintain a required separation from a lead object and/or rear road user, i.e. a required headway, in accordance with adaptive cruise control methodologies.

The machine learning algorithm is for controlling one or more parameters of one or more of the other algorithms, in dependence on information indicative of past use of the host vehicle 10. The information may be indicative of past use of the host vehicle 10 in the autonomous mode and/or the non-autonomous mode. The information may be indicative of inputs such as steering inputs, acceleration inputs and braking inputs. The information may be indicative of environment characteristics. The information may be associated with information from the sensing means. The information may be associated with traffic conditions, road works or the like. The information may be indicative of locations of the past use. The information may be indicative of a temporal pattern of use of the host vehicle 10. For example, the times of the past use may have been recorded. The temporal pattern may enable locations visited at a recurring time and/or day and/or date to be established, such as a workplace. The information may be used for training of the machine learning algorithm. Machine learning enables an optimization of vehicle behavior for repeated journeys. Further, at least some of the parameters may be user-settable using HMI according to preference.

Whether a manoeuvre is performed may be subject to one or more manoeuvring constraints. If a manoeuvring constraint cannot be satisfied, the path planning algorithm may need to modify the manoeuvre or an abort condition for aborting the manoeuvre may even be satisfied. In an example, the abort condition may be satisfied when the cost of performing the manoeuvre is high. The abort condition may be satisfied when the cost of performing the manoeuvre is higher or a threshold amount higher than the cost of performing a different manoeuvre. If an abort condition is satisfied, the manoeuvre is not performed. The abort conditions may be checked just before performing the manoeuvre. An example check for satisfaction of the abort condition comprises continually detecting objects as described above. An object may render an intended manoeuvre or already planned path inappropriate.

A static object obstructing the path may be such an object. Examples include roadworks or debris intersecting the path. Another road user, whether moving or not, may also render the manoeuvre or path inappropriate. The check may be dependent on an expected trajectory of the other road user relative to the planned path of the host vehicle 10. If the other road user is predicted to need to change its speed and/or course as a result of the manoeuvre by the host vehicle 10, the abort condition may be satisfied. The check may depend on detection of signals of intent from the other road users such as turn signals. If an abort condition is satisfied, the host vehicle 10 may remain in the autonomous mode and the speed and/or path may be re-planned accordingly.

In certain circumstances, the autonomous mode may need to hand control at least partially back to the occupant by switching to the non-autonomous mode. The non-autonomous mode may be entirely non-autonomous or may be less autonomous than the autonomous mode. The non-autonomous mode may require manual control or at least supervision by a human driver. The non-autonomous mode may comprise one or more driver assistance functions. For example, the non-autonomous mode may comprise at least one of the following functions: cruise control; adaptive cruise control; lane keeping assistance; braking assistance; overtaking assistance; parking assistance.

The control system 200 may be configured to receive at least one further signal indicative of a requirement to switch from the autonomous mode to the non-autonomous mode. The further signal may be indicative of a vehicle characteristic. The further signal may be indicative of a user characteristic. The further signal may be indicative of an environment characteristic. The further signal may be from the sensing means, or from another part of the control system 200 such as an algorithm that processes the map data and/or dynamic data.

The control system 200 may be configured to cause output of a user prompt signal in dependence on the further signal, for example if it is determined that a required highway exit junction approached by the host vehicle 10 is within a threshold driving time and/or distance. The user prompt signal may prompt the occupant to take an action to enable the host vehicle 10 to transition out of the autonomous mode. If the occupant takes the prompted action, the host vehicle 10 transitions out of the autonomous mode. lithe occupant does not take the prompted action, the occupant may be determined to be non-responsive which is an internal hazard associated with the host vehicle 10, therefore the control system 200 may determine a requirement to stop the host vehicle 10 and cause the host vehicle 10 to stop accordingly. In some examples, the requirement to stop may be determined before the user prompt signal is output, for example in dependence on a vehicle characteristic, user characteristic and/or environment characteristic. For example, a failure of a vehicle component may have occurred or the occupant may be unconscious.

The user prompt signal may be presented to the occupant via HMI. The control system 200 may be configured to receive a user readiness signal from the occupant in response to the user prompt signal. The user readiness signal may be transmitted in dependence on user actuation of HMI or a vehicle control such as the steering wheel. In one example, the HMI comprises a plurality of input HMIs on the steering wheel. The input HMI may comprise buttons or any other appropriate means. The input HMIs may be located at left and right sides of the steering wheel with reference to a centred steering wheel, i.e. no steering lock applied. The input HMIs may be located such that at least one digit of each the occupant's hands can remain at least partially hooked over the circumferential tube-like member of the steering wheel, at 9 o'clock and 3 o'clock or 10 o'clock and 2 o'clock positions, when the input HMIs are actuated by the occupant's hands. The input HMIs may need to be pressed concurrently and/or for a threshold duration.

Additionally or alternatively, the user readiness signal may be transmitted in dependence on user actuation of a vehicle control such as the steering wheel, accelerator pedal or brake pedal. For example, turning the steering wheel or depressing the pedal by more than a threshold amount causes the user readiness signal to be transmitted. In other examples, the HMI could take any other appropriate form.

The control system 200 may be configured to determine whether a user readiness signal has been received within a predetermined period of time from the user prompt signal. For example, the predetermined period of time may be from the range approximately 10 seconds to several minutes, depending on the required trade-off between user reaction time and maximum autonomous mode driving time. If the autonomous mode is for highway driving only, the predetermined period of time may be longer, in the order of minutes rather than seconds. For example, the predetermined period of time may be two or more minutes. The predetermined period of time may depend on the level of autonomy of the host vehicle 10, and may be greater for level four than for level three. The predetermined period of time may vary in use in dependence upon the vehicle characteristic, the user characteristic and/or the environment characteristic. The predetermined period of time may be settable by the occupant although may not be below a minimum time. The control system 200 may be configured to output one or more reminder signals for presentation to the occupant via HMI, between transmitting the user prompt signal and receiving the user readiness signal. For example, the reminder signal may comprise at least one of an audible alert, a haptic alert, a visual alert. A characteristic of the reminder signals such as a frequency, volume, number of output HMIs employed, may vary for each subsequent reminder signal. In an implementation, the user prompt signal at 0 seconds causes an audible instruction, a first reminder signal at 20 seconds causes another audible instruction, and subsequent reminder signals at 30, 40, 50 seconds etc., each cause a combination of an audible instruction, and hapfic pulses through the driver's seat and/or steering wheel.

The environment characteristic, vehicle characteristic and/or user characteristic may be as described above, wherein the user prompt signal is transmitted if one or more of the criteria described above are no longer satisfied. Additionally or alternatively, different environment characteristics, vehicle characteristics and/or user characteristics may be defined for the determination whether to transmit the user prompt signal.

Regarding the environment characteristic, the control system 200 may be configured to transmit the user prompt signal in response to a current or upcoming change of driving environment. The upcoming change may be within a threshold distance or driving time. The change may be caused by non-satisfaction of the road type criterion and/or weather criterion as described above. Additionally or alternatively the change may be caused by detection of one or more of: a traffic light on the road; a toll booth on the road; an off-ramp from the road for following a navigation route. The off-ramp may specifically be for leaving a highway onto a minor road, rather than for transifioning from one highway to another highway.

Regarding the user characteristic, the control system 200 may be configured to transmit the user prompt signal in response to a changed user characteristic. For example, the change may be caused by non-satisfaction of the awareness criterion and/or the physiological criterion. In an implementation, the user prompt signal may be transmitted if the occupant is drowsy or unconscious.

Regarding the vehicle characteristic, the control system 200 may be configured to transmit the user prompt signal in response to a changed vehicle characteristic. For example, the change may be caused by non-satisfaction of the detection range criterion or any other of the criterion or situations described earlier. Additionally or alternatively, the change may be caused by a determination of a fault with the host vehicle 10, which is defined as a type of internal hazard associated with the host vehicle 10. The fault may be caused by one or more of: power failure; communication failure; or sensing means failure. The power failure may comprise an electrical power failure such as a failure of the power supply and/or backup power supply. The power failure may comprise a mechanical power failure such as an inhibited availability of propulsive torque from the prime mover, which may be caused by the prime mover becoming inoperable or entering a limp home mode. The mechanical power failure may correspond to a failure of a drivetrain component such as the transmission or differential. The mechanical power failure may correspond to a failure of an actuator with a responsibility for the dynamic driving task in autonomous mode. The power failure may comprise failure of headlamps at night. The communication failure may comprise a failure of one or more of the electronic communication networks. The communication failure may comprise a failure of one or more controllers with a responsibility for the dynamic driving task in autonomous mode. The communication failure may comprise a failure of a domain controller. The sensing means failure may comprise a failure of one or more of the sensing means. The fault may trigger a determination that the host vehicle 10 is to stop. The user prompt signal may be transmitted to enable the occupant to control how the host vehicle 10 is stopped. The control system 200 may be configured to stop the host vehicle 10 without driver intervention.

Various methods will now be described for being performed during autonomous driving in the autonomous mode. At least some of the methods are in accordance with one or more aspects of the present invention. The control system 200 could be configured to implement one or more of the methods. Computer software could be configured to, when executed, perform one or more of the methods via the control system 200.

With reference to Figs 3 to 5, there is provided a method 2100, 2200 for controlling the host vehicle 10 operable in the autonomous mode (and, in some examples, operable in the non-autonomous mode), the method comprising: determining 2102, 2202 a traffic condition 2000, beyond a first sensing range of the host vehicle 10; associating 2104, 2204 the traffic condition 2000 with a first lane of a multi-lane highway comprising a plurality of lanes in a first direction including the first lane; and causing 2106, 2210 control of a direction of the host vehicle 10 before the host vehicle 10 reaches the location of the traffic condition 2000 to change lane from one of the plurality of lanes to another of the plurality of lanes of the multilane highway, in dependence on the traffic condition 2000 being associated with the first lane.

By way of context, on highways road users will try to change lanes earlier than they normally would to avoid or anticipate a traffic condition as soon as the driver has line of sight to the traffic condition and becomes aware of it. Often, they will increase their speed upon changing lanes, which unfortunately boxes in other road users in front of them who may wish to perform the same manoeuvre.

Vehicles in an autonomous mode are reliant on sensing means for performing a driving task and may not be capable of an earlier-than-normal lane change, causing them to react to the traffic condition later than human drivers.

Sometimes, autonomous vehicles may be in the optimum lane for a traffic condition, but they may not stay in the optimum lane. They may automatically leave the lane to overtake another road user, causing the vehicle to 'lose its place' in the traffic. The vehicle may have to try and push back in, disrupting traffic flow. Consequently, autonomous vehicles may get stuck in traffic and/or boxed in frequently and may disrupt the flow of traffic. The methods 2100 and 2200 disclosed herein alleviate these problems.

Fig 3 illustrates an example operating scenario. The host vehicle 10 is in the autonomous mode and is on a highway. In Fig 3 there are multiple available lanes for travel in the intended direction. A traffic condition 2000 is ahead of the host vehicle 10 in the same lane as the host vehicle 10. In other examples, the traffic condition may be in a different lane from the host vehicle 10. The traffic condition 2000 is beyond the first sensing range so is undetectable by the sensing means, and the traffic condition 2000 may also be beyond a line of sight of the occupant.

In some examples, the traffic condition 2000 may comprise a slower-moving or stationary traffic queue. In some examples, the traffic queue may be entering the lane, such as from a highway on-ramp. In some examples, the traffic condition 2000 such as the traffic queue may be associated with a junction. For example, the traffic queue may be caused by road users entering the lane from the highway on-ramp or the traffic queue may be for exiting the highway at a highway off-ramp. A traffic condition 2000 may not be determined if there is no traffic queue associated with the junction.

In some examples, the traffic condition 2000 comprises a closure of a lane, with or without a traffic queue. There may be no choice but to exit the closed lane.

Fig 4 illustrates an example method 2100. At block 2102, the method comprises determining the traffic condition 2000. The traffic condition 2000 is beyond the first sensing range of the host vehicle 10. The traffic condition 2000 may be determined by processing any appropriate information that indicates traffic conditions beyond the first sensing range. For example, the information may indicate traffic conditions within the second farther sensing range. The information may be dynamic. The information may comprise the map data and/or dynamic data as described earlier. The information may comprise information associated with traffic from V2V or V2I communication. The dynamic data may be from a remote information source 302 as described earlier. The dynamic data may be substantially live and may be updated periodically as described earlier.

The dynamic map data may comprise information on lane closures as previously described. The dynamic traffic data may comprise information on traffic conditions such as traffic queues as previously described. The map data and/or dynamic map data may comprise information on traffic conditions such as lane closures. The location of the traffic condition may be determined. The road signs may be permanent road signs. In some examples, road sign information comprises temporary signs such as lane closure warning signs, which are typically temporary and occur during roadworks and/or at certain times of day or in response to hazardous obstructions such as crashed vehicles.

The map data may comprise information on locations associated with junctions such as junction locations, on-ramp locations, off-ramp locations, or a combination thereof. The map data may indicate whether the on-ramp merges with an existing lane of the highway. The map data may indicate whether the on-ramp continues as an extra lane on the highway. The map data may indicate which existing lane of the highway becomes a lane of the off-ramp.

One of the above-described traffic conditions is a traffic queue. Not all traffic queues may be determined to be traffic conditions in block 2102. For example, the traffic queue may need to satisfy a severity condition to be determined as a traffic condition. Satisfaction could be determined in one or more ways by the host vehicle 10 or the provider of the dynamic traffic data. For example, the traffic speed or flow may need to be below a threshold. The traffic density may need to be above a threshold. Inter-vehicular distance may need to be below a threshold. The threshold(s) may need to be exceeded for more than a threshold distance (e.g. traffic queue length) and/or frequency. Travelling between two points on the highway may need to take an above-threshold journey time. A spatial and/or temporal change in one or more of these variables may need to indicate worsening conditions rather than improving conditions. For instance, traffic density ahead of the host vehicle 10 may need to be increasing relative to a current traffic density, rather than falling. Checking the above variable(s) ensures that a traffic queue is distinguished from temporary and normal fluctuations in traffic flow. This reduces the likelihood of false positives.

At block 2104, the method comprises associating the traffic condition 2000 with a first lane of the highway. The information may have sufficient resolution, granularity and/or detail to enable the traffic condition 2000 to be associated with a specific lane of a highway, as described earlier. Similarly, the information may be substantially live and updated frequently as described earlier. The first lane may or may not be the lane on which the host vehicle 10 is planned to travel by the path planning algorithm.

The traffic condition 2000 may need to be associated with fewer than all lanes in the first direction. In the case of lane closures, the information may indicate this directly by indicating which lanes are closed and/or which are open.

In the case of traffic queues, the control system 200 or provider of the dynamic data may be configured to determine a relative severity of the traffic condition 2000 associated with the first lane compared to other lanes in the first direction. If the difference in severity is not significant, no association with the first lane may be made and the method may terminate. For example, if a difference in severity is above a threshold for a subset of lanes to indicate increased traffic/queuing in the subset of lanes, the traffic condition 2000 may be associated with the subset of lanes. Alternatively or additionally, if the severity condition is satisfied for a subset of the lanes and not others, then the traffic condition 2000 may be associated with the subset of lanes.

In some examples, the method may further comprise associating the traffic condition 2000 with a junction. For example, the method may comprise determining the location associated with the junction. The method may compare the location associated with the junction with one or more locations associated with the traffic condition 2000. The method may determine that the traffic condition 2000, such as a traffic queue, starts before and ends at or not after the location associated with the junction. This association between the locations enables a determination to be made that the traffic queue is associated with the junction. When such an association is made, block 2204 may associate the traffic condition 2000 with the lane(s) determined to facilitate access to/from the junction. In the case of an off-ramp, the determination may be that a traffic queue in the first lane is for leaving the highway at the junction. In the case of an on-ramp, the determination may be that the traffic queue is on the first lane for allowing traffic on the on-ramp to enter the first lane, or even that the traffic queue is on the on-ramp itself.

At block 2106, the method comprises causing control of a direction of the host vehicle 10 before the host vehicle 10 reaches the location of the traffic condition 2000, in dependence on the traffic condition 2000 being associated with the first lane.

The control may cause the host vehicle 10 to change lane from one of the plurality of lanes to another of the plurality of lanes. In some use cases, the lane change may be to increase a separation from the first lane where the traffic condition 2000 is. The host vehicle 10 may change into a lane that is expected to have better traffic flow. In other examples, if the traffic condition 2000 is for leaving the highway at a junction and the navigation route leaves the highway at the junction, the lane change may be to get into the first lane or at least change lanes towards the first lane. The control may be performed by one or more of the output signals described above. In some situations, a lane change may not be necessary. For example, the host vehicle 10 may already be in the most advantageous lane.

The lane change may be performed earlier than normal. The lane change may be performed at a greater time and/or distance from the traffic condition 2000 than a lane change associated with the vehicle overtaking function.

In some examples, a parameter of the nearside bias function may be changed, such as the threshold time of the nearside bias function described earlier. The threshold time could be extended due to the association. This enables a lane change to an offside lane to be performed earlier than normal. If the lane change is to a nearside lane the lane change may be performed even earlier than the nearside bias function would require.

Additionally or alternatively, the earlier-than-normal lane change may be implemented by changing a parameter of the lane change algorithm. For example, the threshold size of an available space associated with the space availability signal could be varied. The threshold could be lowered to enable a more assertive lane change. It is therefore likely that the lane change will be performed earlier than normal.

In some examples, the earlier-than-normal lane change may be performed in dependence on a specific location of the traffic condition 2000. The specific location could be of the back of a traffic queue, a junction sign, or a lane closure warning sign, depending on the traffic condition 2000. Advantageously, the location could be of the first of several junction signs or the first of several lane closure warning signs, because attentive other road users tend to change lanes once they have seen the first sign rather than once the junction or closure is in sight.

The control may be implemented once the host vehicle 10 is within a threshold proximity (distance/time) to the location.

Once the host vehicle 10 has performed the required lane change, the method may optionally comprise inhibiting one or more parameters of the lane change algorithm, such as the vehicle overtaking function. The nearside bias function may be inhibited if the host vehicle 10 is in an offside lane. For example, the vehicle overtaking function may be inhibited from checking for at least certain opportunities to overtake other road users, if such manoeuvres would cause the host vehicle 10 to lose its advantageous position on the highway. An overtaking manoeuvre back into the lane that the host vehicle 10 just left may be inhibited. Leaving the lane may be inhibited in the situation where the first lane is for a junction off-ramp to be used by the host vehicle 10. Overtaking slower road users could be prohibited altogether, for example when the host vehicle 10 has entered a traffic queue in the first lane for leaving the highway at a junction. The inhibition may be removed once the traffic condition 2000 is passed by the host vehicle 10.

In some examples, the host vehicle 10 may be controlled between block 2102 and block 2106, or between block 2104 and block 2106, to make the lane change easier. For example, the host vehicle 10 may be controlled to seek or maintain a traffic position laterally adjacent a space in a neighbouring lane of an above-threshold size. In some examples, a traffic position may be sought with above-threshold sized spaces in both of the neighbouring lanes to either side of the current lane, if a direction of a lane change has not yet been decided.

Fig 5 illustrates a method 2200. Block 2202 is the same as block 2102. Block 2204 is the same as block 2104. Block 2210 is the same as block 2106. The method 2200 selects a most appropriate lane to change to, at block 2206. The method 2200 reduces the impact of a false positive determination at block 2202/2204 by introducing a decision block 2208.

Block 2206 relates to a specific selection of another lane as a target lane for the lane change. Block 2206 is shown to be performed after block 2204, but not necessarily in all examples.

In a use case in which the traffic condition 2000 is to be avoided, a criterion for selecting a lane may be to reduce travel time. In one example implementation, the method may comprise selecting the target lane in dependence on a determination that traffic density, speed, flow and/or journey time in the another lane is more favourable than in a different one or more of the plurality of lanes such as the lane the host vehicle 10 is currently in or planned to be in.

In a use case in which the traffic condition 2000 is to not be avoided, such as joining a traffic queue for leaving a junction, the method may comprise selecting the target lane in dependence on a navigation constraint. The navigation constraint may be defined by the navigation route. The navigation constraint may require exiting the highway at a predetermined junction. If the traffic condition 2000 is determined to be for leaving the highway at the junction as described above, the selected target lane may be the lane or lanes that enable access to leaving the highway at the junction.

In some examples, the target lane selection may be constrained by one or more other constraints. One constraint may be from the nearside bias function which requires a nearside lane to be selected except for overtaking. Another constraint may be to avoid carpool-only or public transport-only lanes if the host vehicle 10 is not permitted to use such lanes.

Before the method progresses to from block 2206 to block 2208, preparatory steps may be performed. For example, the speed of the host vehicle 10 may be controlled in dependence on the space availability signal as described above, to find opportunities for a lane change.

Indicator lights of the host vehicle 10 may be flashed. The lane position may be biased off-centre in the direction of the target lane.

The method 2200 progresses to block 2208. The lane change (block 2210) is performed in dependence on a confirmation at block 2208 of the traffic condition 2000. In an example, the confirmation may be performed by the host vehicle 10 using information from the sensing means of the host vehicle 10. The concept of verification from the short-range high-trust sensing mode (the sensing means) is described earlier. The verification may comprise, for example, checking the data from the sensing means to look for at least one feature associated with the traffic condition and expected from block 2202. The feature may define a trigger condition for the lane change. Example features are described above, such as the presence of the back of a traffic queue, a first lane closure warning sign, or a first junction sign.

In some examples, a plurality of said features and/or a change in at least one of the features is assessed by trend analysis. For example, falling traffic speed, rising density, or a combination thereof may be determined by trend analysis. This enables an early confirmation that the traffic condition exists, before the host vehicle 10 is sufficiently close to the traffic condition that satisfaction of the severity condition can be determined using the sensing means alone.

Alternatively or additionally to using the sensing means, the confirmation may be performed by the host vehicle 10 using information from communication with another road user proximal to the traffic condition. Such communication may be V2V. The another road user may be proximal to the traffic condition wherein the another road user comprises sensing means and the traffic condition is within the first sensing range of the sensing means of the another road user.

The back of the traffic queue may be detected by analysing the dynamics of other road users. For instance, a decreasing speed or decreasing inter-vehicular distance may be observed at around the expected location of the back of the traffic queue, in the first lane.

The lane change may be performed without delay upon verification, such as within about two seconds, to ensure that the host vehicle 10 reacts quickly and before other road users box the host vehicle 10 in.

If the feature associated with the traffic condition is not verified via the sensing means, the method may instead progress to block 2212 in which the host vehicle 10 stays in its current lane when it otherwise would have changed lane. This advantageously means that a false positive determination/association will not result in unexpected vehicle behaviour. This reflects that the dynamic data may not be a reliable indicator of traffic conditions compared to the sensing means. At block 2212, any lane changes are performed in the regular non-early manner, unaffected by the earlier determination and association steps.

Alternatively or in addition to performing blocks 2204 and 2206 before block 2208, blocks 2204 and 2206 may occur between blocks 2208 and 2210. This is to provide greater certainty, of which lane(s) are associated with the traffic condition and/or which lane to select. This is because the sensing means or V2V information from block 2208 may provide greater certainty that the dynamic data from block 2202.

With reference to Figs 6 and 7, there is provided a method 11000 for controlling the host vehicle 10 operable in the autonomous mode (and, in some examples, operable in the non-autonomous mode), the method comprising: determining 11002 a traffic disruption; associating 11004 the traffic disruption with a first lane; and causing 11006 control of a direction of the host vehicle 10 to cause the host vehicle 10 to increase a lateral offset of the host vehicle 10 from a centre of a current lane of the host vehicle 10, wherein the current lane is the same as or is adjacent to the first lane, wherein the lateral offset is in a direction away from the first lane, while remaining within the current lane, in dependence on the associating.

A typical lane centring algorithm may implement a function such as a cost map that favours a centred lane position. If the cost of adjusting lane position to be off-centre is favourable compared to being centred, then the predetermined lateral position of the host vehicle 10 may change to the lower cost off-centre position. The cost map may be changed in dependence on side separation from other road users or the lane change algorithm. However, it would be beneficial if the lane centring algorithm could adjust vehicle position pre-emptively in dependence on traffic disruptions, to increase its lateral separation.

At block 11002, the method comprises determining a traffic disruption. A traffic disruption may impede progress of vehicles at the location of the traffic disruption. For example, the traffic disruption may comprise a traffic condition such as a traffic queue, and/or a closed lane. The traffic disruption may impede traffic flow or speed. The traffic disruption may be associated with a plurality of other road users, for example the traffic disruption may comprise a traffic queue.

However, the term traffic disruption is not limited to such situations. Traffic disruptions include, for example, various environments associated with a high speed difference and a small lateral separation from other road users. For example, the traffic disruption may be associated with passing or being passed by an object at a high speed with a high speed difference. Such environments may not necessarily be an impediment to traffic flow or speed. However, such environments may disrupt visibility of the road ahead, and/or may not leave a wide lateral safety margin. See for example the 'trajectory of a usual autonomous car' in Fig 6, which stays in the centre of a lane adjacent to a lane of stopped traffic queuing for an off-ramp, leaving only a narrow lateral safety margin.

Examples of traffic disruptions include one or more of: a traffic condition; a junction; temporary street furniture (e.g. traffic cones); closure of a lane; a vehicle filtering (lane splitting) through traffic such as a motorcycle or pushbike, or an emergency services vehicle; or a wide load vehicle. The traffic condition may comprise a slow-moving or stopped traffic queue or a closed lane. In some, but not necessarily all examples, the junction, temporary street furniture and/or lane closure may only be a traffic disruption if associated with a traffic queue at or approaching its location.

The traffic disruption may be beyond the first sensing range of the host vehicle 10 or within the first sensing range. For example, the information may indicate traffic disruptions beyond the first sensing range and within the second farther sensing range. The information may comprise the map data and/or dynamic data as described earlier.

In some examples, the information for determining the traffic disruption in block 11002 may comprise one or more of the following earlier described features: the map data, comprising information on the junction; the road sign information, comprising information on the junction and/or the closure of a lane; the dynamic map data, comprising information on the closure of a lane; or the dynamic traffic data, comprising information on the traffic condition. This enables traffic disruptions beyond the first sensing range to be determined.

In some examples, the information utilised in block 11002 may comprise one or more of: information from the sensing means (particularly one or more front and/or rear sensors) on the host vehicle 10 with line of sight to the traffic disruption and/or one or more other road users.

Less predictable traffic disruptions such as wide load vehicles and filtering road users may be detected using the sensing means. The rear sensor may detect filtering road users. A road user may be determined as a filtering road user in dependence on at least one of: a position of the filtering road user; a classification of the filtering road user; a swarm behaviour of traffic around the host vehicle 10. The position may comprise a position straddling multiple lanes. The classification may recognise the road user as an emergency services vehicle and/or a classification that a light signal corresponds to that of an emergency services vehicle. The classification may recognise motorbikes and/or push bikes. The swarm behaviour may comprise road users around the host vehicle 10 adjusting their lateral positions to create a corridor for the filtering road user.

A road user may be determined as a wide load vehicle in dependence on a width of the road user. If the width is above a threshold, the road user may be a wide load vehicle. The front sensor could detect wide load vehicles.

The wide load vehicle and/or filtering vehicle may communicate its presence to the host vehicle 10 via vehicle-to-vehicle communication.

The road sign information may be derived from information sensed by the sensing means, or from the map data and/or dynamic map data comprising the road sign information. Metadata or image data may indicate the purpose and/or any markings on the road signs, for interpretation by the control system 200. The road signs may be permanent road signs. In some examples, the dynamic map data may comprise road sign information to enable temporary signs such as lane closure warning signs to be determined, which are typically temporary and occur during roadworks and/or at certain times of day or in response to hazardous obstructions such as crashed vehicles.

The information on the junctions may comprise information on locations associated with junctions such as junction locations, on-ramp locations, off-ramp locations, or a combination thereof. The map data may indicate whether traffic on the on-ramp has to merge with traffic in an existing lane of the highway. The map data may indicate whether the on-ramp continues as an extra lane on the highway. Likewise, the map data may indicate whether traffic has to leave an existing lane to use an off-ramp. The map data may indicate whether an existing lane of the highway becomes a lane of the off-ramp. These indications may be distinguishable from the detail of the map data such as road markings.

One of the above-described traffic conditions is a traffic queue. Not all traffic queues may be determined to be traffic conditions in block 2102. For example, the traffic queue may need to satisfy a severity condition to be determined as a traffic condition. Satisfaction could be determined in one or more ways by the host vehicle 10 or the provider of the dynamic traffic data. For example, the traffic speed may need to be below a threshold. The traffic density may need to be above a threshold. Inter-vehicular distance may need to be below a threshold. The threshold(s) may need to be exceeded for more than a threshold distance (e.g. traffic queue length) and/or frequency. Travelling between two points on the highway may need to take an above-threshold journey time. A spatial and/or temporal change in one or more of these variables may need to indicate worsening conditions rather than improving conditions. For instance, traffic density ahead of the host vehicle 10 may need to be increasing relative to a current traffic density, rather than falling. Checking the above variable(s) ensures that a traffic queue is distinguished from temporary and normal fluctuations in traffic flow. This reduces the likelihood of false positives.

The temporary street furniture may be detected directly using the sensing means, or indirectly using the dynamic data. The temporary street furniture may comprise traffic cones or temporary barriers, for example. For example, the dynamic map data may indicate that a lane closure is associated with roadworks which may be necessary and sufficient for a detection of street furniture.

In some examples, the traffic disruption may be detected by other means, such as V2V or V2I communication with the host vehicle 10.

At block 11004, the method comprises associating the traffic disruption with a first lane. The first lane may be a lane of a plurality of lanes for travel in a first direction, on the road or highway on which the host vehicle 10 is travelling. The information may have sufficient resolution, granularity and/or detail to enable the traffic disruption to be associated with a specific lane of a highway, as described earlier. Similarly, the information may be substantially live and updated frequently as described earlier. The first lane may or may not be the lane on which the host vehicle 10 is planned to travel by the path planning algorithm.

The traffic disruption may be associated with the first lane specifically, as opposed to all lanes in the first direction. In the case of lane closures, the information may indicate this directly by indicating which lanes are closed and/or which are open.

In the case of traffic queues, the control system 200 or provider of the dynamic data may be configured to determine a relative severity of the traffic disruption associated with the first lane compared to other lanes in the first direction. If the difference in severity is not significant, no association with the first lane may be made and the method may terminate. For example, the relative severity may be determined by determining which lanes satisfy the severity disruption and/or by comparing the magnitudes of variables as disclosed above in relation to block 2102. For example, if a difference in magnitudes is above a threshold for a subset of lanes to indicate increased traffic/queuing in the subset of lanes, the traffic disruption may be associated with the subset of lanes. Alternatively or additionally, if the severity disruption is satisfied for a subset of the lanes and not others, then the traffic disruption may be associated with the subset of lanes.

In some examples, the method may further comprise associating the traffic disruption with a junction. For example, the method may comprise determining the location associated with the junction. The method may compare the location associated with the junction with one or more locations associated with the traffic disruption. The method may determine that the traffic disruption, such as a traffic queue, starts before and ends at or not after the location associated with the junction. This association between the locations enables a determination to be made that the traffic queue is associated with the junction. When such an association is made, block 11004 may associate the traffic disruption with the lane(s) determined to meet the junction to facilitate access to/from the junction. In the case of an off-ramp, the determination may be that a traffic queue in the first lane is for leaving the highway at the junction. In the case of an on-ramp, the determination may be that the traffic queue is on the first lane for allowing traffic on the on-ramp to enter the first lane, or even that the traffic queue is on the on-ramp itself.

Associating temporary street furniture with the first lane may comprise determining that the temporary street furniture is at least on a lane boundary of the first lane, optionally protruding into the first lane beyond the lane boundary. This may be detected directly using the sensing means. However, in some examples, this may be detected by associating the roadworks with a lane directly neighbouring the first lane. Then it may be assumed that temporary street furniture is present.

Associating a filtering road user with the first lane may be dependent on one or more of: the position of the filtering road user; the swarm behaviour. One of the lanes that is being straddled may be the first lane. The swarm behaviour may comprise creating a corridor for the filtering vehicle that is at least partially in the first lane.

Associating a wide load vehicle with the first lane may comprise determining an indication that a lateral edge of the wide load is within a threshold proximity of a first side (lateral boundary) of the first lane and/or at least partially intrudes into the first side of the first lane.

At block 11006, the method comprises causing control of a direction of the host vehicle 10 to cause the host vehicle 10 to increase a lateral offset of the host vehicle 10 from a centre of a current lane of the host vehicle 10, in dependence on block 11004. A rate of lateral deviation associated with the increase of lateral offset may depend on a comfort condition. The comfort condition may impose a performance index limiting jerk and/or acceleration.

The current lane may be adjacent to the first lane. This is shown in Fig 6. The increased lateral offset is also illustrated as the 'proposed improved trajectory'. The lateral offset is in a direction away from the first lane. The additional lateral offset from the traffic condition provides various advantages as mentioned above. In the case of Fig 6, if the host vehicle 10 needs to suddenly change its lateral position further, e.g. a vehicle unexpectedly pulls out of the first lane into the current lane, the amount of required lateral deviation, particularly jerk, is reduced because the host vehicle 10 is already laterally offset away from the first lane.

As shown in Fig 6, the lateral offset may be gradually increased as the host vehicle 10 approaches the traffic disruption. The lateral offset increases pre-emptively, before host vehicle 10 and the traffic disruption are at the same location. The lateral offset may then be gradually decreased after the traffic disruption has passed or been passed. The lateral offset may gradually decrease until the traffic disruption-independent lane position is reached (e.g. lane centre).

In some, but not necessarily all examples, the current lane may be the first lane. lithe host vehicle 10 is in the first lane, the host vehicle 10 may plan a lane change which starts with drifting laterally towards a target lane as per block 11006. That way, if the host vehicle 10 has to stop/match traffic disruption speed before a space availability signal has been received, the host vehicle 10 will not have to make large steering inputs to pull out of the first lane and into the target lane.

The host vehicle 10 in its laterally offset position may remain within the boundaries of the current lane, to comply with applicable highway law. Therefore, in some examples the lane change algorithm is not triggered in the method 11000. In some examples, the host vehicle in the laterally offset position may be no closer than approximately 10cm to 20cm to the closest lateral boundary of the current lane while the host vehicle 10 is in the laterally offset position. The range may define a maximum threshold proximity to the lateral boundary, to avoid confusing other road users in an adjacent lane on the other side of the lateral boundary, regarding the intentions of the host vehicle 10. In some examples, the amount of lateral offset may be constrained by a required minimum lateral separation distance from a road user within the adjacent lane. If the road user in the adjacent lane is detected passing or adjacent to the host vehicle 10, the amount of lateral offset may be reduced by the constraint.

The lateral offset may depend on a detected lateral offset of one or more other road users ahead of and/or behind the host vehicle 10. The host vehicle 10 may obey the swarm behaviour of the other road users.

In the example where a road user is filtering through traffic, the direction of the lateral offset may be away from the path of the filtering vehicle. The amount of lateral offset may be controlled to create a corridor of threshold width to allow a filtering vehicle to pass the host vehicle 10. The threshold width may be less for a motorcycle or push bike than for an emergency services vehicle. If the lateral separation of the host vehicle 10 from another road user adjacent the host vehicle 10 exceeds the threshold width, the lateral offset (if any required) may be sufficient.

Implementing block 11006 may comprise modifying the lane centring algorithm. For example, a cost of following a path that is not laterally offset away from the first lane may exceed the cost of following a path that is laterally offset away from the first lane.

Considering the example of a junction, the control system 200 may be configured to determine in dependence on navigation information whether the traffic disruption is to be avoided. The traffic disruption may be a traffic condition such as a traffic queue.

The control system 200 may determine whether the host vehicle 10 is to exit the highway at the junction. The first lane may meet the junction as described above. If the host vehicle 10 is to remain on the highway, the lateral offset may be away from the first lane as described above, such as shown in Fig 6.

If the host vehicle 10 is to exit the highway at the junction and the host vehicle 10 is in the first lane already, no lateral offset may be implemented. In some examples, lateral offset may be increased in the direction of the junction, to signal an intention to leave the highway at the junction. If the host vehicle 10 is to exit the highway at the junction and the host vehicle 10 is not in the first lane, a lane change may be implemented into the first lane. The lane change may be as described in relation to the method 2100, 2200. The lateral offset may be towards the first lane and in advance of the lane change, to signal an intention to leave the highway at the junction.

Considering the example of a filtering vehicle, the control system 200 may be configured to determine if the host vehicle 10 is able to pull over, crossing the lane boundary of the current lane. In some situations, the applicable highway law takes precedence over allowing emergency services vehicles through. For example, the host vehicle 10 may be prohibited from crossing the lateral boundary of the current lane when there is no adequate space for a lane change. The host vehicle 10 may be prohibited from travelling too slowly. The control system 200 may be configured to implement block 11006 if the host vehicle 10 is unable to pull over. If the host vehicle 10 is able to pull over, the lateral offset may be beyond the lateral boundary of the lane and into a verge or adjacent lane.

Considering the example of the lane closure, the offset is in a direction away from the first lane. The increased lateral offset therefore reduces any laterally evasive manoeuvring required by the host vehicle 10, if road users pull out of the first lane just before the lane closure location, leaving little time for reaction.

Considering the example of the temporary street furniture, the offset is in a direction away from the temporary street furniture.

In some examples, the lateral offset of block 11006 may be accompanied by a change of speed of the host vehicle 10. The change of speed may be to reduce a speed difference between the host vehicle 10 and the traffic disruption. The change of speed may comprise deceleration of the host vehicle 10. The deceleration may comprise friction braking, overrun braking and/or regenerative braking. The host vehicle 10 may decelerate to a reduced target speed for passing the traffic disruption. The reduced speed further reduces the sensation of high speed because the host vehicle 10 may still be laterally quite close to the traffic disruption, even at the offset position.

In some jurisdictions and road types, an emergency corridor may need to be created when vehicles stop, even if there is no approaching emergency services vehicle. Therefore, the lateral offset may be temporarily overridden by an emergency corridor function when the host vehicle 10 is stopping, for instance in stop-start traffic.

In some examples, the control system 200 may cause further control of the host vehicle 10 to change lane to a second lane that is to a same side of the centre of the current lane as the lateral offset. The lateral offset may therefore be in preparation for the lane change. The lateral offset may be while the control system 200 waits for the space availability signal which is received when the traffic-aware lane change algorithm determines a threshold space (gap) in traffic in the second lane.

In some examples, block 11006 may be performed subject to confirmation that the traffic disruption exists, from a confirmation block equivalent to block 2208 of the method 2200.

For purposes of this disclosure, it is to be understood that the controller(s) 202 described herein can each comprise a control unit or computational device having one or more electronic processors 204. A vehicle and/or a system thereof may comprise a single control unit or electronic controller or alternatively different functions of the controller(s) may be embodied in, or hosted in, different control units or controllers. A set of instructions 208 could be provided which, when executed, cause said controller(s) or control unit(s) to implement the control techniques described herein (including the described method(s)). The set of instructions may be embedded in one or more electronic processors, or alternatively, the set of instructions could be provided as software to be executed by one or more electronic processor(s). For example, a first controller may be implemented in software run on one or more electronic processors, and one or more other controllers may also be implemented in software run on or more electronic processors, optionally the same one or more processors as the first controller. It will be appreciated, however, that other arrangements are also useful, and therefore, the present disclosure is not intended to be limited to any particular arrangement. In any event, the set of instructions described above may be embedded in a computer-readable storage medium 210 (e.g., a non-transitory computer-readable storage medium) that may comprise any mechanism for storing information in a form readable by a machine or electronic processors/computational device, including, without limitation: a magnetic storage medium (e.g., floppy diskette); optical storage medium (e.g., CD-ROM); magneto optical storage medium; read only memory (ROM); random access memory (RAM); erasable programmable memory (e.g., EPROM ad EEPROM); flash memory; or electrical or other types of medium for storing such information/instructions.

The term 'if' is used herein in relation to the concept of conditional performance of a function if' a condition is satisfied. The term 'if' in this context means that the function is capable of being performed if the condition is satisfied and is not capable of being performed if the condition is not satisfied. Additional conditions (not stated) may also need to be satisfied before the function is performed. Therefore, although it may be that the stated condition is the only condition for performing some functions, the 'if' terminology herein does not limit to such scenarios.

Separation', 'distance' and 'position' as disclosed herein are not intended to be limited to absolute values of distance. The terms can be normalised by speed. For instance, a separation or distance may be two seconds (at 10 metres per second).

The blocks illustrated in the Figures may represent steps in a method and/or sections of code in the computer program 208. The illustration of a particular order to the blocks does not necessarily imply that there is a required or preferred order for the blocks and the order and arrangement of the block may be varied. Furthermore, it may be possible for some steps to be omitted.

Although embodiments of the present invention have been described in the preceding paragraphs with reference to various examples, it should be appreciated that modifications to the examples given can be made without departing from the scope of the invention as claimed. For the absence of doubt, the autonomous mode may be operable in non-highway roads.

Features described in the preceding description may be used in combinations other than the combinations explicitly described.

Although functions have been described with reference to certain features, those functions may be performable by other features whether described or not.

Although features have been described with reference to certain embodiments, those features may also be present in other embodiments whether described or not.

Whilst endeavoring in the foregoing specification to draw attention to those features of the invention believed to be of particular importance it should be understood that the Applicant claims protection in respect of any patentable feature or combination of features hereinbefore referred to and/or shown in the drawings whether or not particular emphasis has been placed thereon.

Claims

CLAIMS1. A control system for a host vehicle operable in an autonomous mode, the control system comprising one or more controllers, the control system configured to: determine a traffic condition, beyond a first sensing range of the host vehicle; associate the traffic condition with a first lane of a multi-lane highway comprising a plurality of lanes in a first direction including the first lane; and cause control of a direction of the host vehicle before the host vehicle reaches the location of the traffic condition to change lane from one of the plurality of lanes to another of the plurality of lanes of the multi-lane highway, in dependence on the traffic condition being associated with the first lane.
2. The control system of claim 1, wherein the one or more controllers collectively comprise: at least one electronic processor having an electrical input for receiving the information; and at least one electronic memory device electrically coupled to the at least one electronic processor and having instructions stored therein; and wherein the at least one electronic processor is configured to access the at least one memory device and execute the instructions thereon so as to cause the host vehicle to perform the determining, the associating and the causing the control.
3. The control system of claim 1 or 2, wherein the lane change is performed at a greater time and/or distance from the traffic condition than a lane change associated with a vehicle overtaking function of the host vehicle.
4. The control system of claim 3, wherein the lane change is farther before the host vehicle reaches the traffic condition for changing from an offside lane to a nearside lane, compared to changing from a nearside lane to an offside lane.
5. The control system of any preceding claim, wherein the control is performed in dependence on confirmation of the traffic condition using information from one or more sensors of the host vehicle having the first sensing range and/or from communication with another road user proximal to the traffic condition.
6. The control system of claim 5, wherein the traffic condition comprises a slower-moving or stationary traffic queue in the first lane.
7. The control system of any preceding claim, wherein the traffic condition is associated with a junction and wherein the control system is configured to determine in dependence on navigation information whether the host vehicle is to exit the multi-lane highway at the junction, wherein if the host vehicle is to exit the multi-lane highway at the junction and the one lane is a second lane different from the first lane, the another lane is the first lane.
8. The control system of any one of claims 1 to 6, wherein the traffic condition comprises a closure of at least one of the plurality of lanes, wherein the associating comprises determining that the closure is of at least the first lane, and wherein if the one lane is the first lane, the another lane is a second lane different from the first lane.
9. The control system of any preceding claim, wherein the determining and/or the associating utilises dynamic information received from a remote information source.
10. The control system of any preceding claim, wherein the determining and/or the associating utilises at least one of: map data comprising information on junctions; road sign information comprising information on junctions; dynamic map data comprising information on lane closures; or dynamic traffic data comprising information on traffic conditions.
11. The control system of any preceding claim, configured to select the another lane from the plurality of lanes, in dependence on one or more of: a determination that traffic speed in the another lane is faster than traffic speed in a different one or more of the plurality of lanes; or a navigation constraint.
12. The control system of any preceding claim, wherein the control of the direction comprises causing a steering subsystem of the host vehicle to control steering of the host vehicle to follow a planned path of the host vehicle determined in dependence on the determining.
13. A method for controlling a host vehicle operable in an autonomous mode, the method comprising: determining a traffic condition, beyond a first sensing range of the host vehicle; associating the traffic condition with a first lane of a multi-lane highway comprising a plurality of lanes in a first direction including the first lane; and causing control of a direction of the host vehicle before the host vehicle reaches the location of the traffic condition to change lane from one of the plurality of lanes to another of the plurality of lanes of the multi-lane highway, in dependence on the traffic condition being associated with the first lane.
14. A vehicle comprising the control system of any of claims 1 to 12.
15. Computer software that, when executed, is arranged to perform a method according to claim 13.
16. A non-transitory, computer-readable storage medium storing instructions thereon that, when executed by one or more electronic processors, causes the one or more electronic processors to carry out the method of claim 13.