Autonomyof vehicles Prof. dr. Jernej Klemenc, dr. Simon Oman 12. Julij 2018
Content Sensors and ICT (information communication technology) Data processing -identification of traffic signalization and various situations 2
Sensors in autonomous vehicle 3
Sensors Lidar sensor Autonomous vehiclesmay use lidarsensorfor obstacle detection and avoidance to navigate safely through environments, using rotating laser beams.cost map or point cloud outputs from the lidarsensor provide the necessary data for robot software to determine where potential obstacles exist in the environment and where the autonomousvehicle is in relation to those potential obstacles. 4
Sensors Radar sensor Radaris an object-detection system that usesradio wavesto determine the range, angle, or velocity of objects. A radar system consists of a transmitter producing electromagnetic waves in the radio or microwaves domain, a transmittingantenna, a receiving antenna (often the same antenna is used for transmitting and receiving) and areceiverandprocessorto determine properties of the object(s). Radio waves (pulsed or continuous) from the transmitter reflect off the object and return to the receiver, giving information about the object's location and 5 speed(doppler radar systems).
Sensors Image sensors Camera Systems using CMOS or CCD image sensors -HD with 2 or more megapixels. Camera using FLIR technology: An infrared camera is a non-contact device that detects infrared energy (heat) and converts it into an electronic signal, which is then processed to producea thermal image on a video monitor 6 and perform temperature calculations.
Sensors Acoustic sensors Acoustic sensor transmitsa pulse of acoustic sound toward a target, which reflects the sound back to the sensor. The system then measures the time for the echo to return to the sensor and computes the distance to the target using the speed of sound in the medium(air). 7
Sensors Example of sensors use active/autonomous cruise control Autonomous cruise control(acc)is an optionalcruise controlsystem forroadvehiclesthat automatically adjusts the vehicle speed to maintain a safe distance from vehicles ahead. Control is based on sensorinformationfromon-boardsensors. Such systems may use aradarorlasersensor or astereo camerasetup allowing the vehicle to brake when it detects the car is approaching another vehicle ahead and accelerates when traffic allows. 8
Information communication technology High-Precision GPS for Autonomous Vehicles A fully autonomous vehicle needs an accurate localization solution paired with the confidence that the localization solution is correct. GNSS technology(receivesgps andglonass signals)is capable of providing decimetre-level accuracy to ensure a vehicle stays in its lane, or a safe distance from other vehicles. 9
Information communication technology Vehicle-to-vehicle communication Vehicle-to-vehicle communication lets cars broadcast their position, speed, steering-wheel position, brake status, and other data to other vehicles within a few hundred meters. The computers aboard each car process the various readings being broadcast (most promising communication technology-new wireless standard, 802.11p) by other vehicles numerus times every second, each time calculating the chance of an impending collision. 10
Data processing To process data received from numerous sensors different expert systems can be used and finally decisions taken: Artificial intelligence(neural networks) Image processing systems 11
Image processing systems Image processing in the car is becoming an increasingly important factor as a driver assistance system. In a combination with theother sensors and powerful pattern recognition methods, it will be a component in the success of autonomous driving. A complete image processing system needs to be capable of: Locating, isolating and monitoring of different objects Recognising located objects Sharingthese data with theother expert systems that can calculate different things and decide what to do Problems: Vast amount of data processing time (on-board or cloud) Huge amount of data and work involved for training the artificialintelligence software. System reliability under non-ideal conditions (rain, fog..) On-board software updates? 12
Communication with theother Anyone who has crossed a busy street likely knows the informal language between pedestrians and drivers. A driver might wave his/her hand to indicate to the pedestrian it s okay to cross, or a pedestrian could throw up his hand like a stop sign to signal he plans to cross first. But what happens in the future, when self-driving vehicles operate without drivers and in some cases, without anyone even in the vehicle itself? visual and audio communication 13
Zebra 14
Steps 15
STOP sign left 16
STOP sign in front 17
Arrows - acceleration 18
Arrows deceleration 19
Pedestrian sign - red 20
Pedestrian sign - green 21
Bar - acceleration 22
Bar - deceleration 23
Bar arrows - acceleration 24
Bar arrows -alert 25
Bar arrows OK to walk 26
Bar arrows deceleration 27
LEDstripe -white 28
LED stripe - acceleration 29
LED stripe OK to walk 30
LED stripe deceleration 1 31
LED stripe deceleration 2 32
LED stripe deceleration 3 33
Examples of possible audio communication: https://www.youtube.com/watch?v=pkd0kn9mxxs https://www.youtube.com/watch?v=xdvsk3-oukg https://www.youtube.com/watch?v=7dsiolstr94 https://www.youtube.com/watch?v=vwffcocwa6u https://www.youtube.com/watch?v=q0cju1ujrwu https://www.youtube.com/watch?v=plixidvgce0 34
A real-life situation to think about it: At one side of the drive lane there is an old person and on the other side there is a mummy with a young baby. An autonomous vehicle correctly recognises all three pedestrians. The autonomous vehicle calculates that it will not be able to stop before the pedestrians. What would be its decision? 35