Scribd
Upload
14 views3 pages

Obstacle Avoidance with Bubble Rebound

The document describes an obstacle avoidance algorithm for a robot using ultrasonic sensors. The robot detects obstacles using a ring of 6 equidistant ultrasonic sensors covering 180 degrees in front. If a sensor detects an obstacle within the robot's "sensitivity bubble", the robot rebounds in a new direction found by calculating a weighted mean of the sensor angles based on their reported distances. This weighted mean direction is the rebound angle, and the robot moves in this new direction until the goal is visible or a new obstacle is encountered.

Uploaded by

Gustavo Da Silva Cruz
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
14 views3 pages

Obstacle Avoidance with Bubble Rebound

The document describes an obstacle avoidance algorithm for a robot using ultrasonic sensors. The robot detects obstacles using a ring of 6 equidistant ultrasonic sensors covering 180 degrees in front. If a sensor detects an obstacle within the robot's "sensitivity bubble", the robot rebounds in a new direction found by calculating a weighted mean of the sensor angles based on their reported distances. This weighted mean direction is the rebound angle, and the robot moves in this new direction until the goal is visible or a new obstacle is encountered.

Uploaded by

Gustavo Da Silva Cruz
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

4.

 Obstacle avoidance via bubble rebound 
algorithm 
4.1 Obstacle detection 
The following reactive algorithm for real time obstacle avoidance is described in [1].  
We consider a robot with a ring of   equidistant ultrasonic sensors in the front, covering an 
angle of 180° as shown below. 
 
Current heading 

A
α 

4  3  B
5  2 

Ultrasonic  beam 
6  1 

Robot 

 
Fig. 4.1.1 Ultrasonic sensor array with sensitivity bubble 
 
 
Let  A  be  the  ultrasonic  range  boundary  (in  our  case  400  cm)  and  B  the  sensitive  bubble 
boundary.  In  the  simplest  case  (if  the  robot  drives  with  invariant  velocity  and  the  time 
intervals  between  successive  evaluations  of  sensor  data  are  equal),  the  distance  of  the 
ultrasonic sensors to the bubble boundary is a constant, called  . If the measured distance 
 of an ultrasonic sensor   to an object is ≤  , the obstacle avoidance algorithm starts, else 
not and the robot continue to drive in current direction. 
 
 
 
 
 
4.2 Obstacle avoidance algorithm 
Initially,  the  robot  moves  straight  towards  the  goal.  If  an  obstacle  is  detected  within  the 
sensitivity bubble, the robots “rebounds” in a direction found as having the lowest density of 
obstacles, and continues its motion in this new direction until the goal becomes visible (i.e. 
no obstacle within the visibility range of the sonar in that direction), or until a new obstacle 
is encountered.  
Since  the  ultrasonic  sensors  are  uniformly  distributed,  covering  an  arc  of  180°,  the  sonar 
readings can be represented in a polar diagram, as shown in figure 4.2.1. 
 
 

α4 

0° 
+18°  ‐18° 
‐54° 
+54° 
4  3 
5  2 
6  i =  1 
+90°  ‐90° 

Sonar readings 

α   
+90°  0°  ‐90° 
 
Fig. 4.2.1 Polar diagram of sonar readings 
 
 
4.3 Computing rebound angle 
The rebound angle is computed by weighted arithmetic mean. The weighted mean is similar 
to the arithmetic mean where instead of each of the data points contributing equally to the final 
average, some data points contribute more than others. Weighted arithmetic mean is computed by 
using following formula: 
∑ ∙
(4.3.1)                ̅ ∑
 
 
where: 
 
̅  is the weighted arithmetic mean 
 is the weight 
 is the data 
 
In our case the weighted arithmetic mean  is the rebound angle  , the data are the sonar 
sensor angles   and the weights are the distance values  , reported by the sonar sensors. 
Thus 
 
 
∑ ∙
(4.3.2)                ∑
 

 
It can be easily noticed that  ,   and so on. Using six sonar sensors as in fig. 
4.2.1 and substituting the sonar sensor angles accordingly, the equation (4.3.2) becomes 
 
 
(4.3.3)    
∙ ∙ ∙ ∙ ∙ ∙
 
 
 
∙ ∙ ∙ ∙ ∙ ∙
 
 
 
∙ ∙ ∙
 
 
To turn the robot precisely into rebound angle position, a compass module will be used. 
 
 
References 
 
[1] IOAN SUSNEA, VIOREL MINZU, GRIGORE VASILIU, " Simple, Real‐Time Obstacle Avoidance 
Algorithm for Mobile Robots", ISBN: 978‐960‐474‐144‐1 

You might also like